Authors: Avi Cooper, Keizo Kato, Chia-Hsien Shih, Hiroaki Yamane, Kasper Vinken, Kentaro Takemoto, Taro Sunagawa, Hao-Wei Yeh, Jin Yamanaka, Ian Mason, Xavier Boix
Abstract: Visual Language Models (VLMs) are now increasingly being merged with Large Language Models (LLMs) to enable new capabilities, particularly in terms of improved interactivity and open-ended responsiveness. While these are remarkable capabilities, the contribution of LLMs to enhancing the longstanding key problem of classifying an image among a set of choices remains unclear. Through extensive experiments involving seven models, ten visual understanding datasets, and multiple prompt variations per dataset, we find that, for object and scene recognition, VLMs that do not leverage LLMs can achieve better performance than VLMs that do. Yet at the same time, leveraging LLMs can improve performance on tasks requiring reasoning and outside knowledge. In response to these challenges, we propose a pragmatic solution: a lightweight fix involving a relatively small LLM that efficiently routes visual tasks to the most suitable model for the task. The LLM router undergoes training using a dataset constructed from more than 2.5 million examples of pairs of visual task and model accuracy. Our results reveal that this lightweight fix surpasses or matches the accuracy of state-of-the-art alternatives, including GPT-4V and HuggingGPT, while improving cost-effectiveness.
Authors: Neal Lawton, Aishwarya Padmakumar, Judith Gaspers, Jack FitzGerald, Anoop Kumar, Greg Ver Steeg, Aram Galstyan
Abstract: QLoRA reduces the memory-cost of fine-tuning a large language model (LLM) with LoRA by quantizing the base LLM. However, quantization introduces quantization errors that negatively impact model performance after fine-tuning. In this paper we introduce QuAILoRA, a quantization-aware initialization for LoRA that mitigates this negative impact by decreasing quantization errors at initialization. Our method spends a small amount of computational overhead to compute this quantization-aware initialization, without increasing the memory-cost of fine-tuning. We evaluate our method on several causal language modeling and downstream evaluation tasks using several different model sizes and families. We observe that almost all LLMs fined-tuned with QuAILoRA achieve better validation perplexity. When evaluated on downstream tasks, we find that QuAILoRA yields improvements proportional to the negative effect of quantization error. On average, applying QuAILoRA to 4-bit QLoRA models yields 75% of the validation perplexity decrease and 86% of the downstream task accuracy increase as doubling the quantization precision to 8-bit, without increasing GPU memory utilization during fine-tuning.
Authors: Fei Liu, Yiming Yao, Ping Guo, Zhiyuan Yang, Xi Lin, Xialiang Tong, Mingxuan Yuan, Zhichao Lu, Zhenkun Wang, Qingfu Zhang
Abstract: Algorithm Design (AD) is crucial for effective problem-solving across various domains. The advent of Large Language Models (LLMs) has notably enhanced the automation and innovation within this field, offering new perspectives and superior solutions. Over the past three years, the integration of LLMs into AD (LLM4AD) has progressed significantly, finding applications in diverse areas such as optimization, machine learning, mathematical reasoning, and scientific exploration. Given the rapid development and broadening scope of this field, a systematic review is both timely and essential. This paper provides a systematic review of the works on LLM4AD. First, we present an overview and summary of existing studies. Then, we present a systematic categorization, and a review of existing works along four dimensions including the role of LLMs, search techniques, prompt strategies, and application fields. We also discuss the achievements and challenges in each area and the capabilities of LLM4AD in addressing them. Finally, we explore current limitations and propose several open questions and promising directions for future research.
Authors: Yuqi Li, Yao Lu, Zeyu Dong, Chuanguang Yang, Yihao Chen, Jianping Gou
Abstract: The deployment of Deep Neural Network (DNN)-based networks on resource-constrained devices remains a significant challenge due to their high computational and parameter requirements. To solve this problem, layer pruning has emerged as a potent approach to reduce network size and improve computational efficiency. However, existing layer pruning methods mostly overlook the intrinsic connections and inter-dependencies between different layers within complicated deep neural networks. This oversight can result in pruned models that do not preserve the essential characteristics of the pre-trained network as effectively as desired. To address this limitations, we propose a Similarity Guided fast Layer Partition pruning for compressing large deep models (SGLP), which focuses on pruning layers from network segments partitioned via representation similarity. Specifically, our presented method first leverages Centered Kernel Alignment (CKA) to indicate the internal representations among the layers of the pre-trained network, which provides us with a potent basis for layer pruning. Based on similarity matrix derived from CKA, we employ Fisher Optimal Segmentation to partition the network into multiple segments, which provides a basis for removing the layers in a segment-wise manner. In addition, our method innovatively adopts GradNorm for segment-wise layer importance evaluation, eliminating the need for extensive fine-tuning, and finally prunes the unimportant layers to obtain a compact network. Experimental results in image classification and for large language models (LLMs) demonstrate that our proposed SGLP outperforms the state-of-the-art methods in both accuracy and computational efficiency, presenting a more effective solution for deploying DNNs on resource-limited platforms. Our codes are available at https://github.com/itsnotacie/information-fusion-SGLP.
URLs: https://github.com/itsnotacie/information-fusion-SGLP.
Authors: Jaime Lien, Laura I. Galindez Olascoaga, Hasan Dogan, Nicholas Gillian, Brandon Barbello, Leonardo Giusti, Ivan Poupyrev
Abstract: The objective of this work is to develop an AI foundation model for physical signals that can generalize across diverse phenomena, domains, applications, and sensing apparatuses. We propose a phenomenological approach and framework for creating and validating such AI foundation models. Based on this framework, we developed and trained a model on 0.59 billion samples of cross-modal sensor measurements, ranging from electrical current to fluid flow to optical sensors. Notably, no prior knowledge of physical laws or inductive biases were introduced into the model. Through several real-world experiments, we demonstrate that a single foundation model could effectively encode and predict physical behaviors, such as mechanical motion and thermodynamics, including phenomena not seen in training. The model also scales across physical processes of varying complexity, from tracking the trajectory of a simple spring-mass system to forecasting large electrical grid dynamics. This work highlights the potential of building a unified AI foundation model for diverse physical world processes.
Authors: Zheng Zhan, Yushu Wu, Zhenglun Kong, Changdi Yang, Yifan Gong, Xuan Shen, Xue Lin, Pu Zhao, Yanzhi Wang
Abstract: Recent advancements in State Space Models (SSMs) have attracted significant interest, particularly in models optimized for parallel training and handling long-range dependencies. Architectures like Mamba have scaled to billions of parameters with selective SSM. To facilitate broader applications using Mamba, exploring its efficiency is crucial. While token reduction techniques offer a straightforward post-training strategy, we find that applying existing methods directly to SSMs leads to substantial performance drops. Through insightful analysis, we identify the reasons for this failure and the limitations of current techniques. In response, we propose a tailored, unified post-training token reduction method for SSMs. Our approach integrates token importance and similarity, thus taking advantage of both pruning and merging, to devise a fine-grained intra-layer token reduction strategy. Extensive experiments show that our method improves the average accuracy by 5.7% to 13.1% on six benchmarks with Mamba-2 compared to existing methods, while significantly reducing computational demands and memory requirements.
Authors: Xiannan Huang, Shuhan Qiu, Yan Cheng, Quan Yuan, Chao Yang
Abstract: Short-term traffic volume prediction is crucial for intelligent transportation system and there are many researches focusing on this field. However, most of these existing researches concentrated on refining model architecture and ignored amount of training data. Therefore, there remains a noticeable gap in thoroughly exploring the effect of augmented dataset, especially extensive historical data in training. In this research, two datasets containing taxi and bike usage spanning over eight years in New York were used to test such effects. Experiments were conducted to assess the precision of models trained with data in the most recent 12, 24, 48, and 96 months. It was found that the training set encompassing 96 months, at times, resulted in diminished accuracy, which might be owing to disparities between historical traffic patterns and present ones. An analysis was subsequently undertaken to discern potential sources of inconsistent patterns, which may include both covariate shift and concept shift. To address these shifts, we proposed an innovative approach that aligns covariate distributions using a weighting scheme to manage covariate shift, coupled with an environment aware learning method to tackle the concept shift. Experiments based on real word datasets demonstrate the effectiveness of our method which can significantly decrease testing errors and ensure an improvement in accuracy when training with large-scale historical data. As far as we know, this work is the first attempt to assess the impact of contiguously expanding training dataset on the accuracy of traffic prediction models. Besides, our training method is able to be incorporated into most existing short-term traffic prediction models and make them more suitable for long term historical training dataset.
Authors: Xiannan Huang, Chao Yang, Quan Yuan
Abstract: Accurate short-term passenger flow prediction of subway stations plays a vital role in enabling subway station personnel to proactively address changes in passenger volume. Despite existing literature in this field, there is a lack of research on effectively integrating features from different periods, particularly intra-period and inter-period features, for subway station passenger flow prediction. In this paper, we propose a novel model called \textbf{M}uti \textbf{P}eriod \textbf{S}patial \textbf{T}emporal \textbf{N}etwork \textbf{MPSTN}) that leverages features from different periods by transforming one-dimensional time series data into two-dimensional matrices based on periods. The folded matrices exhibit structural characteristics similar to images, enabling the utilization of image processing techniques, specifically convolutional neural networks (CNNs), to integrate features from different periods. Therefore, our MPSTN model incorporates a CNN module to extract temporal information from different periods and a graph neural network (GNN) module to integrate spatial information from different stations. We compared our approach with various state-of-the-art methods for spatiotemporal data prediction using a publicly available dataset and achieved minimal prediction errors. The code for our model is publicly available in the following repository: https://github.com/xiannanhuang/MPSTN
Authors: Dana Weitzner, Mauricio Delbracio, Peyman Milanfar, Raja Giryes
Abstract: Recently, diffusion models have gained popularity due to their impressive generative abilities. These models learn the implicit distribution given by the training dataset, and sample new data by transforming random noise through the reverse process, which can be thought of as gradual denoising. In this work, we examine the generation process as a ``correlation machine'', where random noise is repeatedly enhanced in correlation with the implicit given distribution. To this end, we explore the linear case, where the optimal denoiser in the MSE sense is known to be the PCA projection. This enables us to connect the theory of diffusion models to the spiked covariance model, where the dependence of the denoiser on the noise level and the amount of training data can be expressed analytically, in the rank-1 case. In a series of numerical experiments, we extend this result to general low rank data, and show that low frequencies emerge earlier in the generation process, where the denoising basis vectors are more aligned to the true data with a rate depending on their eigenvalues. This model allows us to show that the linear diffusion model converges in mean to the leading eigenvector of the underlying data, similarly to the prevalent power iteration method. Finally, we empirically demonstrate the applicability of our findings beyond the linear case, in the Jacobians of a deep, non-linear denoiser, used in general image generation tasks.
Authors: Bokai Lin, Zihao Zeng, Zipeng Xiao, Siqi Kou, Tianqi Hou, Xiaofeng Gao, Hao Zhang, Zhijie Deng
Abstract: KV cache has become a de facto technique for the inference of large language models (LLMs), where tensors of shape (layer number, head number, sequence length, feature dimension) are introduced to cache historical information for self-attention. As the size of the model and data grows, the KV cache can quickly become a bottleneck within the system in both storage and memory transfer. To address this, prior studies usually focus on the first three axes of the cache tensors for compression. This paper supplements them, focusing on the feature dimension axis, by utilizing low-rank projection matrices to transform the cache features into spaces with reduced dimensions. We begin by investigating the canonical orthogonal projection method for data compression through principal component analysis (PCA). We observe the issue with PCA projection where significant performance degradation is observed at low compression rates. To bridge the gap, we propose to directly tune the orthogonal projection matrices with a distillation objective using an elaborate Matryoshka training strategy. After training, we adaptively search for the optimal compression rates for various layers and heads given varying compression budgets. Compared to previous works, our method can easily embrace pre-trained LLMs and hold a smooth tradeoff between performance and compression rate. We empirically witness the high data efficiency of our training procedure and find that our method can sustain over 90% performance with an average KV cache compression rate of 60% (and up to 75% in certain extreme scenarios) for popular LLMs like LLaMA2-7B-base and Mistral-7B-v0.3-base.
Authors: Jingyi Xu, Yeqi Luo, Weidong Yang, Keyi Liu, Shengnan Wang, Ben Fei, Lei Bai
Abstract: Arctic sea ice performs a vital role in global climate and has paramount impacts on both polar ecosystems and coastal communities. In the last few years, multiple deep learning based pan-Arctic sea ice concentration (SIC) forecasting methods have emerged and showcased superior performance over physics-based dynamical models. However, previous methods forecast SIC at a fixed temporal granularity, e.g. sub-seasonal or seasonal, thus only leveraging inter-granularity information and overlooking the plentiful inter-granularity correlations. SIC at various temporal granularities exhibits cumulative effects and are naturally consistent, with short-term fluctuations potentially impacting long-term trends and long-term trends provides effective hints for facilitating short-term forecasts in Arctic sea ice. Therefore, in this study, we propose to cultivate temporal multi-granularity that naturally derived from Arctic sea ice reanalysis data and provide a unified perspective for modeling SIC via our Sea Ice Foundation Model. SIFM is delicately designed to leverage both intra-granularity and inter-granularity information for capturing granularity-consistent representations that promote forecasting skills. Our extensive experiments show that SIFM outperforms off-the-shelf deep learning models for their specific temporal granularity.
Authors: Guanglin Niu
Abstract: Knowledge Graph Embedding (KGE) techniques play a pivotal role in transforming symbolic Knowledge Graphs (KGs) into numerical representations, thereby enhancing various deep learning models for knowledge-augmented applications. Unlike entities, relations in KGs are the carriers of semantic meaning, and their accurate modeling is crucial for the performance of KGE models. Firstly, we address the complex mapping properties inherent in relations, such as one-to-one, one-to-many, many-to-one, and many-to-many mappings. We provide a comprehensive summary of relation-aware mapping-based models, models that utilize specific representation spaces, tensor decomposition-based models, and neural network-based models. Next, focusing on capturing various relation patterns like symmetry, asymmetry, inversion, and composition, we review models that employ modified tensor decomposition, those based on modified relation-aware mappings, and those that leverage rotation operations. Subsequently, considering the implicit hierarchical relations among entities, we introduce models that incorporate auxiliary information, models based on hyperbolic spaces, and those that utilize the polar coordinate system. Finally, in response to more complex scenarios such as sparse and dynamic KGs, this paper discusses potential future research directions. We explore innovative ideas such as integrating multimodal information into KGE, enhancing relation pattern modeling with rules, and developing models to capture relation characteristics in dynamic KGE settings.
Authors: Balaji Shesharao Ingole, Vishnu Ramineni, Nikhil Bangad, Koushik Kumar Ganeeb, Priyankkumar Patel
Abstract: The primary aim of this paper is to comprehend, assess, and analyze the role, relevance, and efficiency of machine learning models in predicting heart disease risks using clinical data. While the importance of heart disease risk prediction cannot be overstated, the application of machine learning (ML) in identifying and evaluating the impact of various features on the classification of patients with and without heart disease, as well as in generating a reliable clinical dataset, is equally significant. This study relies primarily on cross-sectional clinical data. The ML approach is designed to enhance the consideration of various clinical features in the heart disease prognosis process. Some features emerge as strong predictors, adding significant value. The paper evaluates seven ML classifiers: Logistic Regression, Random Forest, Decision Tree, Naive Bayes, k-Nearest Neighbors, Neural Networks, and Support Vector Machine (SVM). The performance of each model is assessed based on accuracy metrics. Notably, the Support Vector Machine (SVM) demonstrates the highest accuracy at 91.51%, confirming its superiority among the evaluated models in terms of predictive capability. The overall findings of this research highlight the advantages of advanced computational methodologies in the evaluation, prediction, improvement, and management of cardiovascular risks. In other words, the strong performance of the SVM model illustrates its applicability and value in clinical settings, paving the way for further advancements in personalized medicine and healthcare.
Authors: Jie Peng, Zhang Cao, Huaizhi Qu, Zhengyu Zhang, Chang Guo, Yanyong Zhang, Zhichao Zhang, Tianlong Chen
Abstract: Although Large Language Models (LLMs) have demonstrated remarkable capabilities, their massive parameter counts and associated extensive computing make LLMs' deployment the main part of carbon emission from nowadays AI applications. Compared to modern GPUs like H$100$, it would be significantly carbon-sustainable if we could leverage old-fashioned GPUs such as M$40$ (as shown in Figure~\ref{fig:tisser}, M$40$ only has one third carbon emission of H$100$'s) for LLM servings. However, the limited High Bandwidth Memory (HBM) available on such GPU often cannot support the loading of LLMs due to the gigantic model size and intermediate activation data, making their serving challenging. For instance, a LLaMA2 model with $70$B parameters typically requires $128$GB for inference, which substantially surpasses $24$GB HBM in a $3090$ GPU and remains infeasible even considering the additional $64$GB DRAM. To address this challenge, this paper proposes a mixed-precision with a model modularization algorithm to enable LLM inference on outdated hardware with resource constraints. (The precision denotes the numerical precision like FP16, INT8, INT4) and multi-level caching (M2Cache).) Specifically, our M2Cache first modulizes neurons in LLM and creates their importance ranking. Then, it adopts a dynamic sparse mixed-precision quantization mechanism in weight space to reduce computational demands and communication overhead at each decoding step. It collectively lowers the operational carbon emissions associated with LLM inference. Moreover, M2Cache introduces a three-level cache management system with HBM, DRAM, and SSDs that complements the dynamic sparse mixed-precision inference. To enhance communication efficiency, M2Cache maintains a neuron-level mixed-precision LRU cache in HBM, a larger layer-aware cache in DRAM, and a full model in SSD.
Authors: Zao Zhang, Huaming Chen, Pei Ning, Nan Yang, Dong Yuan
Abstract: In knowledge distillation, a primary focus has been on transforming and balancing multiple distillation components. In this work, we emphasize the importance of thoroughly examining each distillation component, as we observe that not all elements are equally crucial. From this perspective,we decouple the Kullback-Leibler (KL) divergence into three unique elements: Binary Classification Divergence (BCD), Strong Correlation Divergence (SCD), and Weak Correlation Divergence (WCD). Each of these elements presents varying degrees of influence. Leveraging these insights, we present the Correlation-Aware Knowledge Distillation (CAKD) framework. CAKD is designed to prioritize the facets of the distillation components that have the most substantial influence on predictions, thereby optimizing knowledge transfer from teacher to student models. Our experiments demonstrate that adjusting the effect of each element enhances the effectiveness of knowledge transformation. Furthermore, evidence shows that our novel CAKD framework consistently outperforms the baseline across diverse models and datasets. Our work further highlights the importance and effectiveness of closely examining the impact of different parts of distillation process.
Authors: Zirui Li, Patrick Wolf, Meng Wang
Abstract: Accurate arrival time prediction (ATP) of buses and trams plays a crucial role in public transport operations. Current methods focused on modeling one-dimensional temporal information but overlooked the latent periodic information within time series. Moreover, most studies developed algorithms for ATP based on a single or a few routes of public transport, which reduces the transferability of the prediction models and their applicability in public transport management systems. To this end, this paper proposes \textit{ArrivalNet}, a two-dimensional temporal variation-based multi-step ATP for buses and trams. It decomposes the one-dimensional temporal sequence into intra-periodic and inter-periodic variations, which can be recast into two-dimensional tensors (2D blocks). Each row of a tensor contains the time points within a period, and each column involves the time points at the same intra-periodic index across various periods. The transformed 2D blocks in different frequencies have an image-like feature representation that enables effective learning with computer vision backbones (e.g., convolutional neural network). Drawing on the concept of residual neural network, the 2D block module is designed as a basic module for flexible aggregation. Meanwhile, contextual factors like workdays, peak hours, and intersections, are also utilized in the augmented feature representation to improve the performance of prediction. 125 days of public transport data from Dresden were collected for model training and validation. Experimental results show that the root mean square error, mean absolute error, and mean absolute percentage error of the proposed predictor decrease by at least 6.1\%, 14.7\%, and 34.2\% compared with state-of-the-art baseline methods.
Authors: Lina Gong, Qi Gao, Peng Li, Mingqiang Wei, Fei Wu
Abstract: Current automatic deep learning (i.e., AutoDL) frameworks rely on training feedback from actual runs, which often hinder their ability to provide quick and clear performance predictions for selecting suitable DL systems. To address this issue, we propose EfficientDL, an innovative deep learning board designed for automatic performance prediction and component recommendation. EfficientDL can quickly and precisely recommend twenty-seven system components and predict the performance of DL models without requiring any training feedback. The magic of no training feedback comes from our proposed comprehensive, multi-dimensional, fine-grained system component dataset, which enables us to develop a static performance prediction model and comprehensive optimized component recommendation algorithm (i.e., {\alpha}\b{eta}-BO search), removing the dependency on actually running parameterized models during the traditional optimization search process. The simplicity and power of EfficientDL stem from its compatibility with most DL models. For example, EfficientDL operates seamlessly with mainstream models such as ResNet50, MobileNetV3, EfficientNet-B0, MaxViT-T, Swin-B, and DaViT-T, bringing competitive performance improvements. Besides, experimental results on the CIFAR-10 dataset reveal that EfficientDL outperforms existing AutoML tools in both accuracy and efficiency (approximately 20 times faster along with 1.31% Top-1 accuracy improvement than the cutting-edge methods). Source code, pretrained models, and datasets are available at https://github.com/OpenSELab/EfficientDL.
Authors: Munsif Ali, Leonardo Rossi, Massimo Bertozzi
Abstract: Few-shot and continual learning face two well-known challenges in GANs: overfitting and catastrophic forgetting. Learning new tasks results in catastrophic forgetting in deep learning models. In the case of a few-shot setting, the model learns from a very limited number of samples (e.g. 10 samples), which can lead to overfitting and mode collapse. So, this paper proposes a Continual Few-shot Teacher-Student technique for the generative adversarial network (CFTS-GAN) that considers both challenges together. Our CFTS-GAN uses an adapter module as a student to learn a new task without affecting the previous knowledge. To make the student model efficient in learning new tasks, the knowledge from a teacher model is distilled to the student. In addition, the Cross-Domain Correspondence (CDC) loss is used by both teacher and student to promote diversity and to avoid mode collapse. Moreover, an effective strategy of freezing the discriminator is also utilized for enhancing performance. Qualitative and quantitative results demonstrate more diverse image synthesis and produce qualitative samples comparatively good to very stronger state-of-the-art models.
Authors: Rhui Dih Lee, Laura Wynter
Abstract: We address the question of how to successively add new knowledge to an LLM whilst retaining previously-added knowledge. We consider two settings, semi-cooperative and fully-cooperative. Overall, LoRA performs better in most cases than full-fine tuning of all parameters when both new knowledge acquisition and retention of old, including recent, knowledge are taken into account. In the semi-cooperative setting, where datasets are not available after training, MOE mixing, model merging, and LoRA-based orthogonal subspace sequential learning, using a small weight on the orthogonality term, perform well. In the fully-cooperative setting where datasets remain available, joint training and sequential training with replay are both effective approaches with LoRA training generally preferable to full fine-tuning. The codes needed to reproduce the results are provided in an open source repository.
Authors: Bac Nguyen, and Chieh-Hsin Lai, Yuhta Takida, Naoki Murata, Toshimitsu Uesaka, Stefano Ermon, Yuki Mitsufuji
Abstract: Latent diffusion models have enabled continuous-state diffusion models to handle a variety of datasets, including categorical data. However, most methods rely on fixed pretrained embeddings, limiting the benefits of joint training with the diffusion model. While jointly learning the embedding (via reconstruction loss) and the latent diffusion model (via score matching loss) could enhance performance, our analysis shows that end-to-end training risks embedding collapse, degrading generation quality. To address this issue, we introduce CATDM, a continuous diffusion framework within the embedding space that stabilizes training. We propose a novel objective combining the joint embedding-diffusion variational lower bound with a Consistency-Matching (CM) regularizer, alongside a shifted cosine noise schedule and random dropping strategy. The CM regularizer ensures the recovery of the true data distribution. Experiments on benchmarks show that CATDM mitigates embedding collapse, yielding superior results on FFHQ, LSUN Churches, and LSUN Bedrooms. In particular, CATDM achieves an FID of 6.81 on ImageNet $256\times256$ with 50 steps. It outperforms non-autoregressive models in machine translation and is on a par with previous methods in text generation.
Authors: Sara Yasmine Ouerk, Olivier Vo Van, Mouadh Yagoubi
Abstract: Predictive maintenance of railway infrastructure, especially railroads, is essential to ensure safety. However, accurate prediction of crack evolution represents a major challenge due to the complex interactions between intrinsic and external factors, as well as measurement uncertainties. Effective modeling requires a multidimensional approach and a comprehensive understanding of these dynamics and uncertainties. Motivated by an industrial use case based on collected real data containing measured crack lengths, this paper introduces a robust Bayesian multi-horizon approach for predicting the temporal evolution of crack lengths on rails. This model captures the intricate interplay between various factors influencing crack growth. Additionally, the Bayesian approach quantifies both epistemic and aleatoric uncertainties, providing a confidence interval around predictions. To enhance the model's reliability for railroad maintenance, specific constraints are incorporated. These constraints limit non-physical crack propagation behavior and prioritize safety. The findings reveal a trade-off between prediction accuracy and constraint compliance, highlighting the nuanced decision-making process in model training. This study offers insights into advanced predictive modeling for dynamic temporal forecasting, particularly in railway maintenance, with potential applications in other domains.
Authors: Amanda A. Howard, Bruno Jacob, Panos Stinis
Abstract: We develop a method for multifidelity Kolmogorov-Arnold networks (KANs), which use a low-fidelity model along with a small amount of high-fidelity data to train a model for the high-fidelity data accurately. Multifidelity KANs (MFKANs) reduce the amount of expensive high-fidelity data needed to accurately train a KAN by exploiting the correlations between the low- and high-fidelity data to give accurate and robust predictions in the absence of a large high-fidelity dataset. In addition, we show that multifidelity KANs can be used to increase the accuracy of physics-informed KANs (PIKANs), without the use of training data.
Authors: Masaru Isonuma, Ivan Titov
Abstract: Fine-tuning is widely used to adapt language models for specific goals, often leveraging real-world data such as patient records, customer-service interactions, or web content in languages not covered in pre-training. These datasets are typically massive, noisy, and often confidential, making their direct inspection challenging. However, understanding them is essential for guiding model deployment and informing decisions about data cleaning or suppressing any harmful behaviors learned during fine-tuning. In this study, we introduce the task of novelty discovery through generation, which aims to identify novel properties of a fine-tuning dataset by generating examples that illustrate these properties. Our approach, Contrastive Generative Exploration (CGE), assumes no direct access to the data but instead relies on a pre-trained model and the same model after fine-tuning. By contrasting the predictions of these two models, CGE can generate examples that highlight novel characteristics of the fine-tuning data. However, this simple approach may produce examples that are too similar to one another, failing to capture the full range of novel phenomena present in the dataset. We address this by introducing an iterative version of CGE, where the previously generated examples are used to update the pre-trained model, and this updated model is then contrasted with the fully fine-tuned model to generate the next example, promoting diversity in the generated outputs. Our experiments demonstrate the effectiveness of CGE in detecting novel content, such as toxic language, as well as new natural and programming languages. Furthermore, we show that CGE remains effective even when models are fine-tuned using differential privacy techniques.
Authors: Bingcong Li, Liang Zhang, Niao He
Abstract: Sharpness-aware minimization (SAM) improves generalization of various deep learning tasks. Motivated by popular architectures such as LoRA, we explore the implicit regularization of SAM for scale-invariant problems involving two groups of variables. Instead of focusing on commonly used sharpness, this work introduces a concept termed balancedness, defined as the difference between the squared norm of two variables. This allows us to depict richer global behaviors of SAM. In particular, our theoretical and empirical findings reveal that i) SAM promotes balancedness; and ii) the regularization on balancedness is data-responsive -- outliers have stronger impact. The latter coincides with empirical observations that SAM outperforms SGD in the presence of outliers. Leveraging the implicit regularization, we develop a resource-efficient SAM variant, balancedness-aware regularization (BAR), tailored for scale-invariant problems such as finetuning language models with LoRA. BAR saves 95% computational overhead of SAM, with enhanced test performance across various tasks on RoBERTa, GPT2, and OPT-1.3B.
Authors: Taiyi Wang, Zhihao Wu, Jianheng Liu, Jianye Hao, Jun Wang, Kun Shao
Abstract: On-device control agents, especially on mobile devices, are responsible for operating mobile devices to fulfill users' requests, enabling seamless and intuitive interactions. Integrating Multimodal Large Language Models (MLLMs) into these agents enhances their ability to understand and execute complex commands, thereby improving user experience. However, fine-tuning MLLMs for on-device control presents significant challenges due to limited data availability and inefficient online training processes. This paper introduces DistRL, a novel framework designed to enhance the efficiency of online RL fine-tuning for mobile device control agents. DistRL employs centralized training and decentralized data acquisition to ensure efficient fine-tuning in the context of dynamic online interactions. Additionally, the framework is backed by our tailor-made RL algorithm, which effectively balances exploration with the prioritized utilization of collected data to ensure stable and robust training. Our experiments show that, on average, DistRL delivers a 3X improvement in training efficiency and enables training data collection 2.4X faster than the leading synchronous multi-machine methods. Notably, after training, DistRL achieves a 20% relative improvement in success rate compared to state-of-the-art methods on general Android tasks from an open benchmark, significantly outperforming existing approaches while maintaining the same training time. These results validate DistRL as a scalable and efficient solution, offering substantial improvements in both training efficiency and agent performance for real-world, in-the-wild device control tasks.
Authors: Hong Jun Jeon, Benjamin Van Roy
Abstract: The staggering feats of AI systems have brought to attention the topic of AI Alignment: aligning a "superintelligent" AI agent's actions with humanity's interests. Many existing frameworks/algorithms in alignment study the problem on a myopic horizon or study learning from human feedback in isolation, relying on the contrived assumption that the agent has already perfectly identified the environment. As a starting point to address these limitations, we define a class of bandit alignment problems as an extension of classic multi-armed bandit problems. A bandit alignment problem involves an agent tasked with maximizing long-run expected reward by interacting with an environment and a human, both involving details/preferences initially unknown to the agent. The reward of actions in the environment depends on both observed outcomes and human preferences. Furthermore, costs are associated with querying the human to learn preferences. Therefore, an effective agent ought to intelligently trade-off exploration (of the environment and human) and exploitation. We study these trade-offs theoretically and empirically in a toy bandit alignment problem which resembles the beta-Bernoulli bandit. We demonstrate while naive exploration algorithms which reflect current practices and even touted algorithms such as Thompson sampling both fail to provide acceptable solutions to this problem, information-directed sampling achieves favorable regret.
Authors: Marco Nurisso, Pierrick Leroy, Francesco Vaccarino
Abstract: Studying the interplay between the geometry of the loss landscape and the optimization trajectories of simple neural networks is a fundamental step for understanding their behavior in more complex settings. This paper reveals the presence of topological obstruction in the loss landscape of shallow ReLU neural networks trained using gradient flow. We discuss how the homogeneous nature of the ReLU activation function constrains the training trajectories to lie on a product of quadric hypersurfaces whose shape depends on the particular initialization of the network's parameters. When the neural network's output is a single scalar, we prove that these quadrics can have multiple connected components, limiting the set of reachable parameters during training. We analytically compute the number of these components and discuss the possibility of mapping one to the other through neuron rescaling and permutation. In this simple setting, we find that the non-connectedness results in a topological obstruction, which, depending on the initialization, can make the global optimum unreachable. We validate this result with numerical experiments.
Authors: Marc A. Tunnell, Zachary J. DeBruine, Erin Carrier
Abstract: Determining the appropriate rank in Non-negative Matrix Factorization (NMF) is a critical challenge that often requires extensive parameter tuning and domain-specific knowledge. Traditional methods for rank determination focus on identifying a single optimal rank, which may not capture the complex structure inherent in real-world datasets. In this study, we introduce a novel approach called Residual Sensitivity to Intial Conditions (RSIC) that suggests potentially multiple ranks of interest by analyzing the sensitivity of the relative residuals (e.g. relative reconstruction error) to different initializations. By computing the Mean Coordinatewise Interquartile Range (MCI) of the residuals across multiple random initializations, our method identifies regions where the NMF solutions are less sensitive to initial conditions and potentially more meaningful. We evaluate RSIC on a diverse set of datasets, including single-cell gene expression data, image data, and text data, and compare it against current state-of-the-art existing rank determination methods. Our experiments demonstrate that RSIC effectively identifies relevant ranks consistent with the underlying structure of the data, outperforming traditional methods in scenarios where they are computationally infeasible or less accurate. This approach provides a more scalable and generalizable solution for rank determination in NMF that does not rely on domain-specific knowledge or assumptions.
Authors: Guangji Bai, Yijiang Li, Zilinghan Li, Liang Zhao, Kibaek Kim
Abstract: Large Language Models (LLMs) achieve state-of-the-art performance but are challenging to deploy due to their high computational and storage demands. Pruning can reduce model size, yet existing methods assume public access to calibration data, which is impractical for privacy-sensitive applications. To address the challenge of pruning LLMs in privacy-preserving settings, we propose FedSpaLLM, the first federated learning framework designed specifically for pruning LLMs. FedSpaLLM enables clients to prune their models locally based on private data while accounting for system heterogeneity and maintaining communication efficiency. Our framework introduces several key innovations: (1) a novel $\ell_0$-norm aggregation function that ensures only non-zero weights are averaged across clients, preserving important model parameters; (2) an adaptive mask expansion technique that meets global sparsity targets while accommodating client-specific pruning decisions; and (3) a layer sampling strategy that reduces communication overhead and personalizes the pruning process based on client resources. Extensive experiments show that FedSpaLLM improves pruning performance in diverse federated settings. The source code will be released upon publication.
Authors: Evan Frick, Tianle Li, Connor Chen, Wei-Lin Chiang, Anastasios N. Angelopoulos, Jiantao Jiao, Banghua Zhu, Joseph E. Gonzalez, Ion Stoica
Abstract: We introduce a new benchmark for reward models that quantifies their ability to produce strong language models through RLHF (Reinforcement Learning from Human Feedback). The gold-standard approach is to run a full RLHF training pipeline and directly probe downstream LLM performance. However, this process is prohibitively expensive. To address this, we build a predictive model of downstream LLM performance by evaluating the reward model on proxy tasks. These proxy tasks consist of a large-scale human preference and a verifiable correctness preference dataset, in which we measure 12 metrics across 12 domains. To investigate which reward model metrics are most correlated to gold-standard RLHF outcomes, we launch an end-to-end RLHF experiment on a large-scale crowdsourced human preference platform to view real reward model downstream performance as ground truth. Ultimately, we compile our data and findings into Preference Proxy Evaluations (PPE), the first reward model benchmark explicitly linked to post-RLHF real-world human preference performance, which we open-source for public use and further development. Our code and evaluations can be found at https://github.com/lmarena/PPE .
Authors: Chaoxi Niu, Hezhe Qiao, Changlu Chen, Ling Chen, Guansong Pang
Abstract: Graph anomaly detection (GAD), which aims to identify nodes in a graph that significantly deviate from normal patterns, plays a crucial role in broad application domains. Existing GAD methods, whether supervised or unsupervised, are one-model-for-one-dataset approaches, i.e., training a separate model for each graph dataset. This limits their applicability in real-world scenarios where training on the target graph data is not possible due to issues like data privacy. To overcome this limitation, we propose a novel zero-shot generalist GAD approach UNPrompt that trains a one-for-all detection model, requiring the training of one GAD model on a single graph dataset and then effectively generalizing to detect anomalies in other graph datasets without any retraining or fine-tuning. The key insight in UNPrompt is that i) the predictability of latent node attributes can serve as a generalized anomaly measure and ii) highly generalized normal and abnormal graph patterns can be learned via latent node attribute prediction in a properly normalized node attribute space. UNPrompt achieves generalist GAD through two main modules: one module aligns the dimensionality and semantics of node attributes across different graphs via coordinate-wise normalization in a projected space, while another module learns generalized neighborhood prompts that support the use of latent node attribute predictability as an anomaly score across different datasets. Extensive experiments on real-world GAD datasets show that UNPrompt significantly outperforms diverse competing methods under the generalist GAD setting, and it also has strong superiority under the one-model-for-one-dataset setting.
Authors: Christopher R. Serrano, Jonathan Gallagher, Kenji Yamada, Alexei Kopylov, Michael A. Warren
Abstract: The boolean satisfiability (SAT) problem asks whether there exists an assignment of boolean values to the variables of an arbitrary boolean formula making the formula evaluate to True. It is well-known that all NP-problems can be coded as SAT problems and therefore SAT is important both practically and theoretically. From both of these perspectives, better understanding the patterns and structure implicit in SAT data is of significant value. In this paper, we describe several advances that we believe will help open the door to such understanding: we introduce hardware accelerated algorithms for fast SAT problem generation, a geometric SAT encoding that enables the use of transformer architectures typically applied to vision tasks, and a simple yet effective technique we term head slicing for reducing sequence length representation inside transformer architectures. These advances allow us to scale our approach to SAT problems with thousands of variables and tens of thousands of clauses. We validate our architecture, termed Satisfiability Transformer (SaT), on the SAT prediction task with data from the SAT Competition (SATComp) 2022 problem sets. Prior related work either leveraged a pure machine learning approach, but could not handle SATComp-sized problems, or was hybrid in the sense of integrating a machine learning component in a standard SAT solving tool. Our pure machine learning approach achieves prediction accuracies comparable to recent work, but on problems that are an order of magnitude larger than previously demonstrated. A fundamental aspect of our work concerns the very nature of SAT data and its suitability for training machine learning models. We both describe experimental results that probe the landscape of where SAT data can be successfully used for learning and position these results within the broader context of complexity and learning.
Authors: Sangyun Lee, Yilun Xu, Tomas Geffner, Giulia Fanti, Karsten Kreis, Arash Vahdat, Weili Nie
Abstract: Consistency models have recently been introduced to accelerate sampling from diffusion models by directly predicting the solution (i.e., data) of the probability flow ODE (PF ODE) from initial noise. However, the training of consistency models requires learning to map all intermediate points along PF ODE trajectories to their corresponding endpoints. This task is much more challenging than the ultimate objective of one-step generation, which only concerns the PF ODE's noise-to-data mapping. We empirically find that this training paradigm limits the one-step generation performance of consistency models. To address this issue, we generalize consistency training to the truncated time range, which allows the model to ignore denoising tasks at earlier time steps and focus its capacity on generation. We propose a new parameterization of the consistency function and a two-stage training procedure that prevents the truncated-time training from collapsing to a trivial solution. Experiments on CIFAR-10 and ImageNet $64\times64$ datasets show that our method achieves better one-step and two-step FIDs than the state-of-the-art consistency models such as iCT-deep, using more than 2$\times$ smaller networks. Project page: https://truncated-cm.github.io/
Authors: Chandrakanth Gudavalli, Bowen Zhang, Connor Levenson, Kin Gwn Lore, B. S. Manjunath
Abstract: In this paper, we present ReeFRAME, a scalable Reeb graph-based framework designed to analyze vast volumes of GPS-enabled human trajectory data generated at 1Hz frequency. ReeFRAME models Patterns-of-life (PoL) at both the population and individual levels, utilizing Multi-Agent Reeb Graphs (MARGs) for population-level patterns and Temporal Reeb Graphs (TERGs) for individual trajectories. The framework's linear algorithmic complexity relative to the number of time points ensures scalability for anomaly detection. We validate ReeFRAME on six large-scale anomaly detection datasets, simulating real-time patterns with up to 500,000 agents over two months.
Authors: I. Luviano Soto, Y. Concha S\'anchez, A. Raya
Abstract: This study investigates the application of an artificial neural network framework for analysing water pollution caused by solids. Water pollution by suspended solids poses significant environmental and health risks. Traditional methods for assessing and predicting pollution levels are often time-consuming and resource-intensive. To address these challenges, we developed a model that leverages a comprehensive dataset of water quality from total suspended solids. A convolutional neural network was trained under a transfer learning approach using data corresponding to different total suspended solids concentrations, with the goal of accurately predicting low, medium and high pollution levels based on various input variables. Our model demonstrated high predictive accuracy, outperforming conventional statistical methods in terms of both speed and reliability. The results suggest that the artificial neural network framework can serve as an effective tool for real-time monitoring and management of water pollution, facilitating proactive decision-making and policy formulation. This approach not only enhances our understanding of pollution dynamics but also underscores the potential of machine learning techniques in environmental science.
Authors: SangJong Lee, Jin-Kwang Kim, JunHo Kim, TaeHan Kim, James Lee
Abstract: In this study, we introduces a parameter-efficient model that outperforms traditional models in time series forecasting, by integrating High-order Polynomial Projection (HiPPO) theory into the Kolmogorov-Arnold network (KAN) framework. This HiPPO-KAN model achieves superior performance on long sequence data without increasing parameter count. Experimental results demonstrate that HiPPO-KAN maintains a constant parameter count while varying window sizes and prediction horizons, in contrast to KAN, whose parameter count increases linearly with window size. Surprisingly, although the HiPPO-KAN model keeps a constant parameter count as increasing window size, it significantly outperforms KAN model at larger window sizes. These results indicate that HiPPO-KAN offers significant parameter efficiency and scalability advantages for time series forecasting. Additionally, we address the lagging problem commonly encountered in time series forecasting models, where predictions fail to promptly capture sudden changes in the data. We achieve this by modifying the loss function to compute the MSE directly on the coefficient vectors in the HiPPO domain. This adjustment effectively resolves the lagging problem, resulting in predictions that closely follow the actual time series data. By incorporating HiPPO theory into KAN, this study showcases an efficient approach for handling long sequences with improved predictive accuracy, offering practical contributions for applications in large-scale time series data.
Authors: Mingan Lin, Fan Yang, Yanjun Shen, Haoze Sun, Tianpeng Li, Tao Zhang, Chenzheng Zhu, Tao Zhang, Miao Zheng, Xu Li, Yijie Zhou, Mingyang Chen, Yanzhao Qin, Youquan Li, Hao Liang, Fei Li, Yadong Li, Mang Wang, Guosheng Dong, Kun Fang, Jianhua Xu, Bin Cui, Wentao Zhang, Zenan Zhou, Weipeng Chen
Abstract: We introduce Baichuan Alignment, a detailed analysis of the alignment techniques employed in the Baichuan series of models. This represents the industry's first comprehensive account of alignment methodologies, offering valuable insights for advancing AI research. We investigate the critical components that enhance model performance during the alignment process, including optimization methods, data strategies, capability enhancements, and evaluation processes. The process spans three key stages: Prompt Augmentation System (PAS), Supervised Fine-Tuning (SFT), and Preference Alignment. The problems encountered, the solutions applied, and the improvements made are thoroughly recorded. Through comparisons across well-established benchmarks, we highlight the technological advancements enabled by Baichuan Alignment. Baichuan-Instruct is an internal model, while Qwen2-Nova-72B and Llama3-PBM-Nova-70B are instruct versions of the Qwen2-72B and Llama-3-70B base models, optimized through Baichuan Alignment. Baichuan-Instruct demonstrates significant improvements in core capabilities, with user experience gains ranging from 17% to 28%, and performs exceptionally well on specialized benchmarks. In open-source benchmark evaluations, both Qwen2-Nova-72B and Llama3-PBM-Nova-70B consistently outperform their respective official instruct versions across nearly all datasets. This report aims to clarify the key technologies behind the alignment process, fostering a deeper understanding within the community. Llama3-PBM-Nova-70B model is available at https://huggingface.co/PKU-Baichuan-MLSystemLab/Llama3-PBM-Nova-70B.
URLs: https://huggingface.co/PKU-Baichuan-MLSystemLab/Llama3-PBM-Nova-70B.
Authors: Hanqun Cao, Chunbin Gu, Mutian He, Ning Ma, Chang-yu Hsieh, Pheng-Ann Heng
Abstract: DNA-encoded library (DEL) screening has revolutionized the detection of protein-ligand interactions through read counts, enabling rapid exploration of vast chemical spaces. However, noise in read counts, stemming from nonspecific interactions, can mislead this exploration process. We present DEL-Ranking, a novel distribution-correction denoising framework that addresses these challenges. Our approach introduces two key innovations: (1) a novel ranking loss that rectifies relative magnitude relationships between read counts, enabling the learning of causal features determining activity levels, and (2) an iterative algorithm employing self-training and consistency loss to establish model coherence between activity label and read count predictions. Furthermore, we contribute three new DEL screening datasets, the first to comprehensively include multi-dimensional molecular representations, protein-ligand enrichment values, and their activity labels. These datasets mitigate data scarcity issues in AI-driven DEL screening research. Rigorous evaluation on diverse DEL datasets demonstrates DEL-Ranking's superior performance across multiple correlation metrics, with significant improvements in binding affinity prediction accuracy. Our model exhibits zero-shot generalization ability across different protein targets and successfully identifies potential motifs determining compound binding affinity. This work advances DEL screening analysis and provides valuable resources for future research in this area.
Authors: Vansh Bansal, Saptarshi Roy, Purnamrita Sarkar, Alessandro Rinaldo
Abstract: Diffusion models have emerged as a powerful tool for image generation and denoising. Typically, generative models learn a trajectory between the starting noise distribution and the target data distribution. Recently Liu et al. (2023b) designed a novel alternative generative model Rectified Flow (RF), which aims to learn straight flow trajectories from noise to data using a sequence of convex optimization problems with close ties to optimal transport. If the trajectory is curved, one must use many Euler discretization steps or novel strategies, such as exponential integrators, to achieve a satisfactory generation quality. In contrast, RF has been shown to theoretically straighten the trajectory through successive rectifications, reducing the number of function evaluations (NFEs) while sampling. It has also been shown empirically that RF may improve the straightness in two rectifications if one can solve the underlying optimization problem within a sufficiently small error. In this paper, we make two key theoretical contributions: 1) we provide the first theoretical analysis of the Wasserstein distance between the sampling distribution of RF and the target distribution. Our error rate is characterized by the number of discretization steps and a new formulation of straightness stronger than that in the original work. 2) In line with the previous empirical findings, we show that, for a rectified flow from a Gaussian to a mixture of two Gaussians, two rectifications are sufficient to achieve a straight flow. Additionally, we also present empirical results on both simulated and real datasets to validate our theoretical findings.
Authors: Zelin Xu, Jie Ren, Yupu Zhang, Jose Maria Gonzalez Ondina, Maitane Olabarrieta, Tingsong Xiao, Wenchong He, Zibo Liu, Shigang Chen, Kaleb Smith, Zhe Jiang
Abstract: Nearly 900 million people live in low-lying coastal zones around the world and bear the brunt of impacts from more frequent and severe hurricanes and storm surges. Oceanographers simulate ocean current circulation along the coasts to develop early warning systems that save lives and prevent loss and damage to property from coastal hazards. Traditionally, such simulations are conducted using coastal ocean circulation models such as the Regional Ocean Modeling System (ROMS), which usually runs on an HPC cluster with multiple CPU cores. However, the process is time-consuming and energy expensive. While coarse-grained ROMS simulations offer faster alternatives, they sacrifice detail and accuracy, particularly in complex coastal environments. Recent advances in deep learning and GPU architecture have enabled the development of faster AI (neural network) surrogates. This paper introduces an AI surrogate based on a 4D Swin Transformer to simulate coastal tidal wave propagation in an estuary for both hindcast and forecast (up to 12 days). Our approach not only accelerates simulations but also incorporates a physics-based constraint to detect and correct inaccurate results, ensuring reliability while minimizing manual intervention. We develop a fully GPU-accelerated workflow, optimizing the model training and inference pipeline on NVIDIA DGX-2 A100 GPUs. Our experiments demonstrate that our AI surrogate reduces the time cost of 12-day forecasting of traditional ROMS simulations from 9,908 seconds (on 512 CPU cores) to 22 seconds (on one A100 GPU), achieving over 450$\times$ speedup while maintaining high-quality simulation results. This work contributes to oceanographic modeling by offering a fast, accurate, and physically consistent alternative to traditional simulation models, particularly for real-time forecasting in rapid disaster response.
Authors: Tianqianjin Lin, Pengwei Yan, Kaisong Song, Zhuoren Jiang, Yangyang Kang, Jun Lin, Weikang Yuan, Junjie Cao, Changlong Sun, Xiaozhong Liu
Abstract: Graph foundation models (GFMs) have recently gained significant attention. However, the unique data processing and evaluation setups employed by different studies hinder a deeper understanding of their progress. Additionally, current research tends to focus on specific subsets of graph learning tasks, such as structural tasks, node-level tasks, or classification tasks. As a result, they often incorporate specialized modules tailored to particular task types, losing their applicability to other graph learning tasks and contradicting the original intent of foundation models to be universal. Therefore, to enhance consistency, coverage, and diversity across domains, tasks, and research interests within the graph learning community in the evaluation of GFMs, we propose GFMBench-a systematic and comprehensive benchmark comprising 26 datasets. Moreover, we introduce LangGFM, a novel GFM that relies entirely on large language models. By revisiting and exploring the effective graph textualization principles, as well as repurposing successful techniques from graph augmentation and graph self-supervised learning within the language space, LangGFM achieves performance on par with or exceeding the state of the art across GFMBench, which can offer us new perspectives, experiences, and baselines to drive forward the evolution of GFMs.
Authors: Yuhao Gong, Yuchen Zhang, Fei Wang, Chi-Han Lee
Abstract: As global climate change intensifies, accurate weather forecasting has become increasingly important, affecting agriculture, energy management, environmental protection, and daily life. This study introduces a hybrid model combining Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks to predict historical temperature data. CNNs are utilized for spatial feature extraction, while LSTMs handle temporal dependencies, resulting in significantly improved prediction accuracy and stability. By using Mean Absolute Error (MAE) as the loss function, the model demonstrates excellent performance in processing complex meteorological data, addressing challenges such as missing data and high-dimensionality. The results show a strong alignment between the prediction curve and test data, validating the model's potential in climate prediction. This study offers valuable insights for fields such as agriculture, energy management, and urban planning, and lays the groundwork for future applications in weather forecasting under the context of global climate change.
Authors: Yunqi Cai, Jiangnan Li, Dong Wang
Abstract: Micromagnetics has made significant strides, particularly due to its wide-ranging applications in magnetic storage design. Numerical simulation is a cornerstone of micromagnetics research, relying on first-principle rules to compute the dynamic evolution of micromagnetic systems based on the renowned LLG equation, named after Landau, Lifshitz, and Gilbert. However, simulations are often hindered by their slow speed. Although Fast-Fourier transformation (FFT) calculations reduce the computational complexity to O(NlogN), it remains impractical for large-scale simulations. In this paper, we introduce NeuralMAG, a deep learning approach to micromagnetic simulation. Our approach follows the LLG iterative framework but accelerates demagnetizing field computation through the employment of a U-shaped neural network (Unet). The Unet architecture comprises an encoder that extracts aggregated spins at various scales and learns the local interaction at each scale, followed by a decoder that accumulates the local interactions at different scales to approximate the global convolution. This divide-and-accumulate scheme achieves a time complexity of O(N), significantly enhancing the speed and feasibility of large-scale simulations. Unlike existing neural methods, NeuralMAG concentrates on the core computation rather than an end-to-end approximation for a specific task, making it inherently generalizable. To validate the new approach, we trained a single model and evaluated it on two micromagnetics tasks with various sample sizes, shapes, and material settings.
Authors: Woojin Chae, Kihyuk Hong, Yufan Zhang, Ambuj Tewari, Dabeen Lee
Abstract: This paper proposes a computationally tractable algorithm for learning infinite-horizon average-reward linear mixture Markov decision processes (MDPs) under the Bellman optimality condition. Our algorithm for linear mixture MDPs achieves a nearly minimax optimal regret upper bound of $\widetilde{\mathcal{O}}(d\sqrt{\mathrm{sp}(v^*)T})$ over $T$ time steps where $\mathrm{sp}(v^*)$ is the span of the optimal bias function $v^*$ and $d$ is the dimension of the feature mapping. Our algorithm applies the recently developed technique of running value iteration on a discounted-reward MDP approximation with clipping by the span. We prove that the value iteration procedure, even with the clipping operation, converges. Moreover, we show that the associated variance term due to random transitions can be bounded even under clipping. Combined with the weighted ridge regression-based parameter estimation scheme, this leads to the nearly minimax optimal regret guarantee.
Authors: Raymundo Vazquez Martinez, Raj Abhijit Dandekar, Rajat Dandekar, Sreedath Panat
Abstract: In this study, we apply two pillars of Scientific Machine Learning: Neural Ordinary Differential Equations (Neural ODEs) and Universal Differential Equations (UDEs) to the Chandrasekhar White Dwarf Equation (CWDE). The CWDE is fundamental for understanding the life cycle of a star, and describes the relationship between the density of the white dwarf and its distance from the center. Despite the rise in Scientific Machine Learning frameworks, very less attention has been paid to the systematic applications of the above SciML pillars on astronomy based ODEs. Through robust modeling in the Julia programming language, we show that both Neural ODEs and UDEs can be used effectively for both prediction as well as forecasting of the CWDE. More importantly, we introduce the forecasting breakdown point - the time at which forecasting fails for both Neural ODEs and UDEs. Through a robust hyperparameter optimization testing, we provide insights on the neural network architecture, activation functions and optimizers which provide the best results. This study provides opens a door to investigate the applicability of Scientific Machine Learning frameworks in forecasting tasks for a wide range of scientific domains.
Authors: Shubhajit Roy, Hrriday Ruparel, Kishan Ved, Anirban Dasgupta
Abstract: Graph Neural Networks (GNNs) have shown remarkable success in various graph-based tasks, including node classification, node regression, graph classification, and graph regression. However, their scalability remains a significant challenge, particularly when dealing with large-scale graphs. To tackle this challenge, coarsening-based methods are used to reduce the graph into a smaller one, resulting in faster computation. However, no previous research has tackled the computation cost during the inference. This motivated us to ponder whether we can trade off the improvement in training time of coarsening-based approaches with inference time. This paper presents a novel approach to improve the scalability of GNNs through subgraph-based techniques. We reduce the computational burden during the training and inference phases by using the coarsening algorithm to partition large graphs into smaller, manageable subgraphs. Previously, graph-level tasks had not been explored using this approach. We propose a novel approach for using the coarsening algorithm for graph-level tasks such as graph classification and graph regression. We conduct extensive experiments on multiple benchmark datasets to evaluate the performance of our approach. The results demonstrate that our subgraph-based GNN method achieves competitive results in node classification, node regression, graph classification, and graph regression tasks compared to traditional GNN models. Furthermore, our approach significantly reduces the inference time, enabling the practical application of GNNs to large-scale graphs.
Authors: Sizhe Liu, Jun Xia, Lecheng Zhang, Yuchen Liu, Yue Liu, Wenjie Du, Zhangyang Gao, Bozhen Hu, Cheng Tan, Hongxin Xiang, Stan Z. Li
Abstract: Molecular relational learning (MRL) is crucial for understanding the interaction behaviors between molecular pairs, a critical aspect of drug discovery and development. However, the large feasible model space of MRL poses significant challenges to benchmarking, and existing MRL frameworks face limitations in flexibility and scope. To address these challenges, avoid repetitive coding efforts, and ensure fair comparison of models, we introduce FlexMol, a comprehensive toolkit designed to facilitate the construction and evaluation of diverse model architectures across various datasets and performance metrics. FlexMol offers a robust suite of preset model components, including 16 drug encoders, 13 protein sequence encoders, 9 protein structure encoders, and 7 interaction layers. With its easy-to-use API and flexibility, FlexMol supports the dynamic construction of over 70, 000 distinct combinations of model architectures. Additionally, we provide detailed benchmark results and code examples to demonstrate FlexMol's effectiveness in simplifying and standardizing MRL model development and comparison.
Authors: Jiahao Wang, Amer Shalaby
Abstract: Accurate prediction of public transit ridership is vital for efficient planning and management of transit in rapidly growing urban areas in Canada. Unexpected increases in passengers can cause overcrowded vehicles, longer boarding times, and service disruptions. Traditional time series models like ARIMA and SARIMA face limitations, particularly in short-term predictions and integration of spatial and temporal features. These models struggle with the dynamic nature of ridership patterns and often ignore spatial correlations between nearby stops. Deep Learning (DL) models present a promising alternative, demonstrating superior performance in short-term prediction tasks by effectively capturing both spatial and temporal features. However, challenges such as dynamic spatial feature extraction, balancing accuracy with computational efficiency, and ensuring scalability remain. This paper introduces DST-TransitNet, a hybrid DL model for system-wide station-level ridership prediction. This proposed model uses graph neural networks (GNN) and recurrent neural networks (RNN) to dynamically integrate the changing temporal and spatial correlations within the stations. The model also employs a precise time series decomposition framework to enhance accuracy and interpretability. Tested on Bogota's BRT system data, with three distinct social scenarios, DST-TransitNet outperformed state-of-the-art models in precision, efficiency and robustness. Meanwhile, it maintains stability over long prediction intervals, demonstrating practical applicability.
Authors: Baojian Zhou, Yifan Sun, Reza Babanezhad Harikandeh, Xingzhi Guo, Deqing Yang, Yanghua Xiao
Abstract: Given the damping factor $\alpha$ and precision tolerance $\epsilon$, \citet{andersen2006local} introduced Approximate Personalized PageRank (APPR), the \textit{de facto local method} for approximating the PPR vector, with runtime bounded by $\Theta(1/(\alpha\epsilon))$ independent of the graph size. Recently, \citet{fountoulakis2022open} asked whether faster local algorithms could be developed using $\tilde{O}(1/(\sqrt{\alpha}\epsilon))$ operations. By noticing that APPR is a local variant of Gauss-Seidel, this paper explores the question of \textit{whether standard iterative solvers can be effectively localized}. We propose to use the \textit{locally evolving set process}, a novel framework to characterize the algorithm locality, and demonstrate that many standard solvers can be effectively localized. Let $\overline{\operatorname{vol}}{ (S_t)}$ and $\overline{\gamma}_{t}$ be the running average of volume and the residual ratio of active nodes $\textstyle S_{t}$ during the process. We show $\overline{\operatorname{vol}}{ (S_t)}/\overline{\gamma}_{t} \leq 1/\epsilon$ and prove APPR admits a new runtime bound $\tilde{O}(\overline{\operatorname{vol}}(S_t)/(\alpha\overline{\gamma}_{t}))$ mirroring the actual performance. Furthermore, when the geometric mean of residual reduction is $\Theta(\sqrt{\alpha})$, then there exists $c \in (0,2)$ such that the local Chebyshev method has runtime $\tilde{O}(\overline{\operatorname{vol}}(S_{t})/(\sqrt{\alpha}(2-c)))$ without the monotonicity assumption. Numerical results confirm the efficiency of this novel framework and show up to a hundredfold speedup over corresponding standard solvers on real-world graphs.
Authors: Mengnan Zhao, Lihe Zhang, Jingwen Ye, Huchuan Lu, Baocai Yin, Xinchao Wang
Abstract: Adversarial training (AT) refers to integrating adversarial examples -- inputs altered with imperceptible perturbations that can significantly impact model predictions -- into the training process. Recent studies have demonstrated the effectiveness of AT in improving the robustness of deep neural networks against diverse adversarial attacks. However, a comprehensive overview of these developments is still missing. This survey addresses this gap by reviewing a broad range of recent and representative studies. Specifically, we first describe the implementation procedures and practical applications of AT, followed by a comprehensive review of AT techniques from three perspectives: data enhancement, network design, and training configurations. Lastly, we discuss common challenges in AT and propose several promising directions for future research.
Authors: Tugrul Cabir Hakyemez, Omer Adar
Abstract: Accurate forecasting of electrical demand is essential for maintaining a stable and reliable power grid, optimizing the allocation of energy resources, and promoting efficient energy consumption practices. This study investigates the effectiveness of five hyperparameter optimization (HPO) algorithms -- Random Search, Covariance Matrix Adaptation Evolution Strategy (CMA--ES), Bayesian Optimization, Partial Swarm Optimization (PSO), and Nevergrad Optimizer (NGOpt) across univariate and multivariate Short-Term Load Forecasting (STLF) tasks. Using the Panama Electricity dataset (n=48,049), we evaluate HPO algorithms' performances on a surrogate forecasting algorithm, XGBoost, in terms of accuracy (i.e., MAPE, $R^2$) and runtime. Performance plots visualize these metrics across varying sample sizes from 1,000 to 20,000, and Kruskal--Wallis tests assess the statistical significance of the performance differences. Results reveal significant runtime advantages for HPO algorithms over Random Search. In univariate models, Bayesian optimization exhibited the lowest accuracy among the tested methods. This study provides valuable insights for optimizing XGBoost in the STLF context and identifies areas for future research.
Authors: Dobrik Georgiev, JJ Wilson, Davide Buffelli, Pietro Li\`o
Abstract: Neural Algorithmic Reasoning (NAR) research has demonstrated that graph neural networks (GNNs) could learn to execute classical algorithms. However, most previous approaches have always used a recurrent architecture, where each iteration of the GNN matches an iteration of the algorithm. In this paper we study neurally solving algorithms from a different perspective: since the algorithm's solution is often an equilibrium, it is possible to find the solution directly by solving an equilibrium equation. Our approach requires no information on the ground-truth number of steps of the algorithm, both during train and test time. Furthermore, the proposed method improves the performance of GNNs on executing algorithms and is a step towards speeding up existing NAR models. Our empirical evidence, leveraging algorithms from the CLRS-30 benchmark, validates that one can train a network to solve algorithmic problems by directly finding the equilibrium. We discuss the practical implementation of such models and propose regularisations to improve the performance of these equilibrium reasoners.
Authors: Keting Yin, Jiayi Mao
Abstract: Federated Learning(FL) is popular as a privacy-preserving machine learning paradigm for generating a single model on decentralized data. However, statistical heterogeneity poses a significant challenge for FL. As a subfield of FL, personalized FL (pFL) has attracted attention for its ability to achieve personalized models that perform well on non-independent and identically distributed (Non-IID) data. However, existing pFL methods are limited in terms of leveraging the global model's knowledge to enhance generalization while achieving personalization on local data. To address this, we proposed a new method personalized Federated learning with Adaptive Feature Aggregation and Knowledge Transfer (FedAFK), to train better feature extractors while balancing generalization and personalization for participating clients, which improves the performance of personalized models on Non-IID data. We conduct extensive experiments on three datasets in two widely-used heterogeneous settings and show the superior performance of our proposed method over thirteen state-of-the-art baselines.
Authors: Ketin Yin, Zonghao Guo, ZhengHan Qin
Abstract: In the current era of artificial intelligence, federated learning has emerged as a novel approach to addressing data privacy concerns inherent in centralized learning paradigms. This decentralized learning model not only mitigates the risk of data breaches but also enhances the system's scalability and robustness. However, this approach introduces a new challenge: how to fairly and accurately assess the contribution of each participant. Developing an effective contribution evaluation mechanism is crucial for federated learning. Such a mechanism incentivizes participants to actively contribute their data and computational resources, thereby improving the overall performance of the federated learning system. By allocating resources and rewards based on the size of the contributions, it ensures that each participant receives fair treatment, fostering sustained engagement.Currently, Shapley value-based methods are widely used to evaluate participants' contributions, with many researchers proposing modifications to adapt these methods to real-world scenarios. In this paper, we introduce a component called Dynamic Pruning Validation Set Shapley (DPVS-Shapley). This method accelerates the contribution assessment process by dynamically pruning the original dataset without compromising the evaluation's accuracy. Furthermore, this component can assign different weights to various samples, thereby allowing clients capable of distinguishing difficult examples to receive higher contribution scores.
Authors: Jiaxuan Gao, Shusheng Xu, Wenjie Ye, Weilin Liu, Chuyi He, Wei Fu, Zhiyu Mei, Guangju Wang, Yi Wu
Abstract: Reward models have been increasingly critical for improving the reasoning capability of LLMs. Existing research has shown that a well-trained reward model can substantially improve model performances at inference time via search. However, the potential of reward models during RL training time still remains largely under-explored. It is currently unclear whether these reward models can provide additional training signals to enhance the reasoning capabilities of LLMs in RL training that uses sparse success rewards, which verify the correctness of solutions. In this work, we evaluate popular reward models for RL training, including the Outcome-supervised Reward Model (ORM) and the Process-supervised Reward Model (PRM), and train a collection of LLMs for math problems using RL by combining these learned rewards with success rewards. Surprisingly, even though these learned reward models have strong inference-time performances, they may NOT help or even hurt RL training, producing worse performances than LLMs trained with the success reward only. Our analysis reveals that an LLM can receive high rewards from some of these reward models by repeating correct but unnecessary reasoning steps, leading to a severe reward hacking issue. Therefore, we introduce two novel reward refinement techniques, including Clipping and Delta. The key idea is to ensure the accumulative reward of any reasoning trajectory is upper-bounded to keep a learned reward model effective without being exploited. We evaluate our techniques with multiple reward models over a set of 1.5B and 7B LLMs on MATH and GSM8K benchmarks and demonstrate that with a carefully designed reward function, RL training without any additional supervised tuning can improve all the evaluated LLMs, including the state-of-the-art 7B LLM Qwen2.5-Math-7B-Instruct on MATH and GSM8K benchmarks.
Authors: Andreas Lang, Erich Schubert
Abstract: The k-means clustering algorithm is a popular algorithm that partitions data into k clusters. There are many improvements to accelerate the standard algorithm. Most current research employs upper and lower bounds on point-to-cluster distances and the triangle inequality to reduce the number of distance computations, with only arrays as underlying data structures. These approaches cannot exploit that nearby points are likely assigned to the same cluster. We propose a new k-means algorithm based on the cover tree index, that has relatively low overhead and performs well, for a wider parameter range, than previous approaches based on the k-d tree. By combining this with upper and lower bounds, as in state-of-the-art approaches, we obtain a hybrid algorithm that combines the benefits of tree aggregation and bounds-based filtering.
Authors: Julian Barra, Shayan Shahbazi, Anthony Birri, Rajni Chahal, Ibrahim Isah, Muhammad Nouman Anwar, Tyler Starkus, Prasanna Balaprakash, Stephen Lam
Abstract: Optimally designing molten salt applications requires knowledge of their thermophysical properties, but existing databases are incomplete, and experiments are challenging. Ideal mixing and Redlich-Kister models are computationally cheap but lack either accuracy or generality. To address this, a transfer learning approach using deep neural networks (DNNs) is proposed, combining Redlich-Kister models, experimental data, and ab initio properties. The approach predicts molten salt density with high accuracy ($r^{2}$ > 0.99, MAPE < 1%), outperforming the alternatives.
Authors: Zixuan Yang, Jiaqi Zheng, Guihai Chen
Abstract: Ensuring verifiable and interpretable safety of deep reinforcement learning (DRL) is crucial for its deployment in real-world applications. Existing approaches like verification-in-the-loop training, however, face challenges such as difficulty in deployment, inefficient training, lack of interpretability, and suboptimal performance in property satisfaction and reward performance. In this work, we propose a novel verification-driven interpretation-in-the-loop framework Reintrainer to develop trustworthy DRL models, which are guaranteed to meet the expected constraint properties. Specifically, in each iteration, this framework measures the gap between the on-training model and predefined properties using formal verification, interprets the contribution of each input feature to the model's output, and then generates the training strategy derived from the on-the-fly measure results, until all predefined properties are proven. Additionally, the low reusability of existing verifiers and interpreters motivates us to develop Reinfier, a general and fundamental tool within Reintrainer for DRL verification and interpretation. Reinfier features breakpoints searching and verification-driven interpretation, associated with a concise constraint-encoding language DRLP. Evaluations demonstrate that Reintrainer outperforms the state-of-the-art on six public benchmarks in both performance and property guarantees. Our framework can be accessed at https://github.com/Kurayuri/Reinfier.
Authors: Haibo Wang, Yuxuan Qiu, Yanze Wang, Rob Brekelmans, Yuanqi Du
Abstract: Simulating transition dynamics between metastable states is a fundamental challenge in dynamical systems and stochastic processes with wide real-world applications in understanding protein folding, chemical reactions and neural activities. However, the computational challenge often lies on sampling exponentially many paths in which only a small fraction ends in the target metastable state due to existence of high energy barriers. To amortize the cost, we propose a data-driven approach to warm-up the simulation by learning nonlinear interpolations from local dynamics. Specifically, we infer a potential energy function from local dynamics data. To find plausible paths between two metastable states, we formulate a generalized flow matching framework that learns a vector field to sample propable paths between the two marginal densities under the learned energy function. Furthermore, we iteratively refine the model by assigning importance weights to the sampled paths and buffering more likely paths for training. We validate the effectiveness of the proposed method to sample probable paths on both synthetic and real-world molecular systems.
Authors: Tianlong Zhou, Jun Shang, Weixiong Rao
Abstract: In this paper, we study the collaborative state fusion problem in a multi-agent environment, where mobile agents collaborate to track movable targets. Due to the limited sensing range and potential errors of on-board sensors, it is necessary to aggregate individual observations to provide target state fusion for better target state estimation. Existing schemes do not perform well due to (1) impractical assumption of the fully known prior target state-space model and (2) observation outliers from individual sensors. To address the issues, we propose a two-stage collaborative fusion framework, namely \underline{L}earnable Weighted R\underline{o}bust \underline{F}usion (\textsf{LoF}). \textsf{LoF} combines a local state estimator (e.g., Kalman Filter) with a learnable weight generator to address the mismatch between the prior state-space model and underlying patterns of moving targets. Moreover, given observation outliers, we develop a time-series soft medoid(TSM) scheme to perform robust fusion. We evaluate \textsf{LoF} in a collaborative detection simulation environment with promising results. In an example setting with 4 agents and 2 targets, \textsf{LoF} leads to a 9.1\% higher fusion gain compared to the state-of-the-art.
Authors: Minhyuk Seo, Hyunseo Koh, Jonghyun Choi
Abstract: The majority of online continual learning (CL) advocates single-epoch training and imposes restrictions on the size of replay memory. However, single-epoch training would incur a different amount of computations per CL algorithm, and the additional storage cost to store logit or model in addition to replay memory is largely ignored in calculating the storage budget. Arguing different computational and storage budgets hinder fair comparison among CL algorithms in practice, we propose to use floating point operations (FLOPs) and total memory size in Byte as a metric for computational and memory budgets, respectively, to compare and develop CL algorithms in the same 'total resource budget.' To improve a CL method in a limited total budget, we propose adaptive layer freezing that does not update the layers for less informative batches to reduce computational costs with a negligible loss of accuracy. In addition, we propose a memory retrieval method that allows the model to learn the same amount of knowledge as using random retrieval in fewer iterations. Empirical validations on the CIFAR-10/100, CLEAR-10/100, and ImageNet-1K datasets demonstrate that the proposed approach outperforms the state-of-the-art methods within the same total budget
Authors: Zhaoxian Wu, Quan Xiao, Tayfun Gokmen, Hsinyu Tsai, Kaoutar El Maghraoui, Tianyi Chen
Abstract: Aiming to accelerate the training of large deep neural models (DNN) in an energy-efficient way, an analog in-memory computing (AIMC) accelerator emerges as a solution with immense potential. In AIMC accelerators, trainable weights are kept in memory without the need to move from memory to processors during the training, reducing a bunch of overhead. However, although the in-memory feature enables efficient computation, it also constrains the use of data parallelism since copying weights from one AIMC to another is expensive. To enable parallel training using AIMC, we propose synchronous and asynchronous pipeline parallelism for AIMC accelerators inspired by the pipeline in digital domains. This paper provides a theoretical convergence guarantee for both synchronous and asynchronous pipelines in terms of both sampling and clock cycle complexity, which is non-trivial since the physical characteristic of AIMC accelerators leads to analog updates that suffer from asymmetric bias. The simulations of training DNN on real datasets verify the efficiency of pipeline training.
Authors: Yinhan He, Zaiyi Zheng, Patrick Soga, Yaozhen Zhu, yushun Dong, Jundong Li
Abstract: In recent years, Graph Neural Networks (GNNs) have become successful in molecular property prediction tasks such as toxicity analysis. However, due to the black-box nature of GNNs, their outputs can be concerning in high-stakes decision-making scenarios, e.g., drug discovery. Facing such an issue, Graph Counterfactual Explanation (GCE) has emerged as a promising approach to improve GNN transparency. However, current GCE methods usually fail to take domain-specific knowledge into consideration, which can result in outputs that are not easily comprehensible by humans. To address this challenge, we propose a novel GCE method, LLM-GCE, to unleash the power of large language models (LLMs) in explaining GNNs for molecular property prediction. Specifically, we utilize an autoencoder to generate the counterfactual graph topology from a set of counterfactual text pairs (CTPs) based on an input graph. Meanwhile, we also incorporate a CTP dynamic feedback module to mitigate LLM hallucination, which provides intermediate feedback derived from the generated counterfactuals as an attempt to give more faithful guidance. Extensive experiments demonstrate the superior performance of LLM-GCE. Our code is released on https://github.com/YinhanHe123/new\_LLM4GNNExplanation.
URLs: https://github.com/YinhanHe123/new\_LLM4GNNExplanation.
Authors: Lincen Bai, Hedi Tabia, Ra\'ul Santos-Rodr\'iguez
Abstract: Deep neural networks (DNNs) excel in tasks like image recognition and natural language processing, but their increasing complexity complicates deployment in resource-constrained environments and increases susceptibility to adversarial attacks. While traditional pruning methods reduce model size, they often compromise the network's ability to withstand subtle perturbations. This paper challenges the conventional emphasis on weight importance scoring as the primary determinant of a pruned network's performance. Through extensive analysis, including experiments conducted on CIFAR, Tiny-ImageNet, and various network architectures, we demonstrate that effective fine-tuning plays a dominant role in enhancing both performance and adversarial robustness, often surpassing the impact of the chosen pruning criteria. To address this issue, we introduce Module Robust Sensitivity, a novel metric that adaptively adjusts the pruning ratio for each network layer based on its sensitivity to adversarial perturbations. By integrating this metric into the pruning process, we develop a stable algorithm that maintains accuracy and robustness simultaneously. Experimental results show that our approach enables the practical deployment of more robust and efficient neural networks.
Authors: Lingyu Zhang, Zhengran Ji, Nicholas R Waytowich, Boyuan Chen
Abstract: The recent rapid advancement of machine learning has been driven by increasingly powerful models with the growing availability of training data and computational resources. However, real-time decision-making tasks with limited time and sparse learning signals remain challenging. One way of improving the learning speed and performance of these agents is to leverage human guidance. In this work, we introduce GUIDE, a framework for real-time human-guided reinforcement learning by enabling continuous human feedback and grounding such feedback into dense rewards to accelerate policy learning. Additionally, our method features a simulated feedback module that learns and replicates human feedback patterns in an online fashion, effectively reducing the need for human input while allowing continual training. We demonstrate the performance of our framework on challenging tasks with sparse rewards and visual observations. Our human study involving 50 subjects offers strong quantitative and qualitative evidence of the effectiveness of our approach. With only 10 minutes of human feedback, our algorithm achieves up to 30% increase in success rate compared to its RL baseline.
Authors: Oussama Boussif, L\'ena N\'ehale Ezzine, Joseph D Viviano, Micha{\l} Koziarski, Moksh Jain, Nikolay Malkin, Emmanuel Bengio, Rim Assouel, Yoshua Bengio
Abstract: As trajectories sampled by policies used by reinforcement learning (RL) and generative flow networks (GFlowNets) grow longer, credit assignment and exploration become more challenging, and the long planning horizon hinders mode discovery and generalization. The challenge is particularly pronounced in entropy-seeking RL methods, such as generative flow networks, where the agent must learn to sample from a structured distribution and discover multiple high-reward states, each of which take many steps to reach. To tackle this challenge, we propose an approach to incorporate the discovery of action abstractions, or high-level actions, into the policy optimization process. Our approach involves iteratively extracting action subsequences commonly used across many high-reward trajectories and `chunking' them into a single action that is added to the action space. In empirical evaluation on synthetic and real-world environments, our approach demonstrates improved sample efficiency performance in discovering diverse high-reward objects, especially on harder exploration problems. We also observe that the abstracted high-order actions are interpretable, capturing the latent structure of the reward landscape of the action space. This work provides a cognitively motivated approach to action abstraction in RL and is the first demonstration of hierarchical planning in amortized sequential sampling.
Authors: Skye Gunasekaran, Assel Kembay, Hugo Ladret, Rui-Jie Zhu, Laurent Perrinet, Omid Kavehei, Jason Eshraghian
Abstract: Accurate time-series forecasting is essential across a multitude of scientific and industrial domains, yet deep learning models often struggle with challenges such as capturing long-term dependencies and adapting to drift in data distributions over time. We introduce Future-Guided Learning, an approach that enhances time-series event forecasting through a dynamic feedback mechanism inspired by predictive coding. Our approach involves two models: a detection model that analyzes future data to identify critical events and a forecasting model that predicts these events based on present data. When discrepancies arise between the forecasting and detection models, the forecasting model undergoes more substantial updates, effectively minimizing surprise and adapting to shifts in the data distribution by aligning its predictions with actual future outcomes. This feedback loop, drawing upon principles of predictive coding, enables the forecasting model to dynamically adjust its parameters, improving accuracy by focusing on features that remain relevant despite changes in the underlying data. We validate our method on a variety of tasks such as seizure prediction in biomedical signal analysis and forecasting in dynamical systems, achieving a 40\% increase in the area under the receiver operating characteristic curve (AUC-ROC) and a 10\% reduction in mean absolute error (MAE), respectively. By incorporating a predictive feedback mechanism that adapts to data distribution drift, Future-Guided Learning offers a promising avenue for advancing time-series forecasting with deep learning.
Authors: Junyang He, Ying-Jung Chen, Anushka Idamekorala, Geoffrey Fox
Abstract: This research is part of a systematic study of scientific time series. In the last three years, hundreds of papers and over fifty new deep-learning models have been described for time series models. These mainly focus on the key aspect of time dependence, whereas in some scientific time series, the situation is more complex with multiple locations, each location having multiple observed and target time-dependent streams and multiple exogenous (known) properties that are either constant or time-dependent. Here, we analyze the hydrology time series using the CAMELS and Caravan global datasets on catchment rainfall and runoff. Together, these have up to 6 observed streams and up to 209 static parameters defined at each of about 8000 locations. This analysis is fully open source with a Jupyter Notebook running on Google Colab for both an LSTM-based analysis and the data engineering preprocessing. Our goal is to investigate the importance of exogenous data, which we look at using eight different choices on representative hydrology tasks. Increasing the exogenous information significantly improves the data representation, with the mean square error decreasing to 60% of its initial value in the largest dataset examined. We present the initial results of studies of other deep-learning neural network architectures where the approaches that can use the full observed and exogenous observations outperform less flexible methods, including Foundation models. Using the natural annual periodic exogenous time series produces the largest impact, but the static and other periodic exogenous streams are also important. Our analysis is intended to be valuable as an educational resource and benchmark.
Authors: Vindula Jayawardana, Baptiste Freydt, Ao Qu, Cameron Hickert, Zhongxia Yan, Cathy Wu
Abstract: Despite the popularity of multi-agent reinforcement learning (RL) in simulated and two-player applications, its success in messy real-world applications has been limited. A key challenge lies in its generalizability across problem variations, a common necessity for many real-world problems. Contextual reinforcement learning (CRL) formalizes learning policies that generalize across problem variations. However, the lack of standardized benchmarks for multi-agent CRL has hindered progress in the field. Such benchmarks are desired to be based on real-world applications to naturally capture the many open challenges of real-world problems that affect generalization. To bridge this gap, we propose IntersectionZoo, a comprehensive benchmark suite for multi-agent CRL through the real-world application of cooperative eco-driving in urban road networks. The task of cooperative eco-driving is to control a fleet of vehicles to reduce fleet-level vehicular emissions. By grounding IntersectionZoo in a real-world application, we naturally capture real-world problem characteristics, such as partial observability and multiple competing objectives. IntersectionZoo is built on data-informed simulations of 16,334 signalized intersections derived from 10 major US cities, modeled in an open-source industry-grade microscopic traffic simulator. By modeling factors affecting vehicular exhaust emissions (e.g., temperature, road conditions, travel demand), IntersectionZoo provides one million data-driven traffic scenarios. Using these traffic scenarios, we benchmark popular multi-agent RL and human-like driving algorithms and demonstrate that the popular multi-agent RL algorithms struggle to generalize in CRL settings.
Authors: Sina Baharlouei, Sadra Sabouri
Abstract: Fair graph clustering is crucial for ensuring equitable representation and treatment of diverse communities in network analysis. Traditional methods often ignore disparities among social, economic, and demographic groups, perpetuating biased outcomes and reinforcing inequalities. This study introduces fair graph clustering within the framework of the disparate impact doctrine, treating it as a joint optimization problem integrating clustering quality and fairness constraints. Given the NP-hard nature of this problem, we employ a semidefinite relaxation approach to approximate the underlying optimization problem. For up to medium-sized graphs, we utilize a singular value decomposition-based algorithm, while for larger graphs, we propose a novel algorithm based on the alternative direction method of multipliers. Unlike existing methods, our formulation allows for tuning the trade-off between clustering quality and fairness. Experimental results on graphs generated from the standard stochastic block model demonstrate the superiority of our approach in achieving an optimal accuracy-fairness trade-off compared to state-of-the-art methods.
Authors: Yujia Wu, Bo Yang, Elynn Chen, Yuzhou Chen, Zheshi Zheng
Abstract: Graph classification in medical imaging and drug discovery requires accuracy and robust uncertainty quantification. To address this need, we introduce Conditional Prediction ROC (CP-ROC) bands, offering uncertainty quantification for ROC curves and robustness to distributional shifts in test data. Although developed for Tensorized Graph Neural Networks (TGNNs), CP-ROC is adaptable to general Graph Neural Networks (GNNs) and other machine learning models. We establish statistically guaranteed coverage for CP-ROC under a local exchangeability condition. This addresses uncertainty challenges for ROC curves under non-iid setting, ensuring reliability when test graph distributions differ from training data. Empirically, to establish local exchangeability for TGNNs, we introduce a data-driven approach to construct local calibration sets for graphs. Comprehensive evaluations show that CP-ROC significantly improves prediction reliability across diverse tasks. This method enhances uncertainty quantification efficiency and reliability for ROC curves, proving valuable for real-world applications with non-iid objects.
Authors: Yujia Wu, Bo Yang, Yang Zhao, Elynn Chen, Yuzhou Chen, Zheshi Zheng
Abstract: Graph Neural Networks (GNNs) have become the de facto standard for analyzing graph-structured data, leveraging message-passing techniques to capture both structural and node feature information. However, recent studies have raised concerns about the statistical reliability of uncertainty estimates produced by GNNs. This paper addresses this crucial challenge by introducing a novel technique for quantifying uncertainty in non-exchangeable graph-structured data, while simultaneously reducing the size of label prediction sets in graph classification tasks. We propose Conformalized Tensor-based Topological Neural Networks (CF-T2NN), a new approach for rigorous prediction inference over graphs. CF-T2NN employs tensor decomposition and topological knowledge learning to navigate and interpret the inherent uncertainty in decision-making processes. This method enables a more nuanced understanding and handling of prediction uncertainties, enhancing the reliability and interpretability of neural network outcomes. Our empirical validation, conducted across 10 real-world datasets, demonstrates the superiority of CF-T2NN over a wide array of state-of-the-art methods on various graph benchmarks. This work not only enhances the GNN framework with robust uncertainty quantification capabilities but also sets a new standard for reliability and precision in graph-structured data analysis.
Authors: Yujia Wu, Junyi Mo, Elynn Chen, Yuzhou Chen
Abstract: Graph contrastive learning (GCL) has emerged as a promising approach to enhance graph neural networks' (GNNs) ability to learn rich representations from unlabeled graph-structured data. However, current GCL models face challenges with computational demands and limited feature utilization, often relying only on basic graph properties like node degrees and edge attributes. This constrains their capacity to fully capture the complex topological characteristics of real-world phenomena represented by graphs. To address these limitations, we propose Tensor-Fused Multi-View Graph Contrastive Learning (TensorMV-GCL), a novel framework that integrates extended persistent homology (EPH) with GCL representations and facilitates multi-scale feature extraction. Our approach uniquely employs tensor aggregation and compression to fuse information from graph and topological features obtained from multiple augmented views of the same graph. By incorporating tensor concatenation and contraction modules, we reduce computational overhead by separating feature tensor aggregation and transformation. Furthermore, we enhance the quality of learned topological features and model robustness through noise-injected EPH. Experiments on molecular, bioinformatic, and social network datasets demonstrate TensorMV-GCL's superiority, outperforming 15 state-of-the-art methods in graph classification tasks across 9 out of 11 benchmarks while achieving comparable results on the remaining two. The code for this paper is publicly available at https://github.com/CS-SAIL/Tensor-MV-GCL.git.
Authors: Shaokang Cheng, Nada Osman, Shiru Qu, Lamberto Ballan
Abstract: High-quality spatiotemporal traffic data is crucial for intelligent transportation systems (ITS) and their data-driven applications. Inevitably, the issue of missing data caused by various disturbances threatens the reliability of data acquisition. Recent studies of diffusion probability models have demonstrated the superiority of deep generative models in imputation tasks by precisely capturing the spatio-temporal correlation of traffic data. One drawback of diffusion models is their slow sampling/denoising process. In this work, we aim to accelerate the imputation process while retaining the performance. We propose a fast conditional diffusion model for spatiotemporal traffic data imputation (FastSTI). To speed up the process yet, obtain better performance, we propose the application of a high-order pseudo-numerical solver. Our method further revs the imputation by introducing a predefined alignment strategy of variance schedule during the sampling process. Evaluating FastSTI on two types of real-world traffic datasets (traffic speed and flow) with different missing data scenarios proves its ability to impute higher-quality samples in only six sampling steps, especially under high missing rates (60\% $\sim$ 90\%). The experimental results illustrate a speed-up of $\textbf{8.3} \times$ faster than the current state-of-the-art model while achieving better performance.
Authors: Haodong Feng, Peiyan Hu, Yue Wang, Dixia Fan
Abstract: In the control problems of the PDE systems, observation is important to make the decision. However, the observation is generally sparse and missing in practice due to the limitation and fault of sensors. The above challenges cause observations with uncertain quantities and modalities. Therefore, how to leverage the uncertain observations as the states in control problems of the PDE systems has become a scientific problem. The dynamics of PDE systems rely on the initial conditions, boundary conditions, and PDE formula. Given the above three elements, PINNs can be used to solve the PDE systems. In this work, we discover that the neural network can also be used to identify and represent the PDE systems using PDE loss and sparse data loss. Inspired by the above discovery, we propose a Physics-Informed Representation (PIR) algorithm for multimodal policies in PDE systems' control. It leverages PDE loss to fit the neural network and data loss calculated on the observations with random quantities and modalities to propagate the information of initial conditions and boundary conditions into the inputs. The inputs can be the learnable parameters or the output of the encoders. Then, under the environments of the PDE systems, such inputs are the representation of the current state. In our experiments, the PIR illustrates the superior consistency with the features of the ground truth compared with baselines, even when there are missing modalities. Furthermore, PIR has been successfully applied in the downstream control tasks where the robot leverages the learned state by PIR faster and more accurately, passing through the complex vortex street from a random starting location to reach a random target.
Authors: Hailiang Zhao, Xueyan Tang, Peng Chen, Shuiguang Deng
Abstract: In this paper, we study learning-augmented algorithms for the Bahncard problem. The Bahncard problem is a generalization of the ski-rental problem, where a traveler needs to irrevocably and repeatedly decide between a cheap short-term solution and an expensive long-term one with an unknown future. Even though the problem is canonical, only a primal-dual-based learning-augmented algorithm was explicitly designed for it. We develop a new learning-augmented algorithm, named PFSUM, that incorporates both history and short-term future to improve online decision making. We derive the competitive ratio of PFSUM as a function of the prediction error and conduct extensive experiments to show that PFSUM outperforms the primal-dual-based algorithm.
Authors: Bo Pan, Zhen Xiong, Guanchen Wu, Zheng Zhang, Yifei Zhang, Liang Zhao
Abstract: Representation learning of Text-Attributed Graphs (TAGs) has garnered significant attention due to its applications in various domains, including recommendation systems and social networks. Despite advancements in TAG learning methodologies, challenges remain in explainability due to the black-box nature of existing TAG representation learning models. This paper presents TAGExplainer, the first method designed to generate natural language explanations for TAG learning. TAGExplainer employs a generative language model that maps input-output pairs to explanations reflecting the model's decision-making process. To address the lack of annotated ground truth explanations in real-world scenarios, we propose first generating pseudo-labels that capture the model's decisions from saliency-based explanations, then the pseudo-label generator is iteratively trained based on three training objectives focusing on faithfulness and brevity via Expert Iteration, to improve the quality of generated pseudo-labels. The high-quality pseudo-labels are finally utilized to train an end-to-end explanation generator model. Extensive experiments are conducted to demonstrate the effectiveness of TAGExplainer in producing faithful and concise natural language explanations.
Authors: Muhammad Aadil Khan, Sai Thatipamula, Simona Onori
Abstract: Real-life batteries tend to experience a range of operating conditions, and undergo degradation due to a combination of both calendar and cycling aging. Onboard health estimation models typically use cycling aging data only, and account for at most one operating condition e.g., temperature, which can limit the accuracy of the models for state-of-health (SOH) estimation. In this paper, we utilize electrochemical impedance spectroscopy (EIS) data from 5 calendar-aged and 17 cycling-aged cells to perform SOH estimation under various operating conditions. The EIS curves are deconvoluted using the distribution of relaxation times (DRT) technique to map them onto a function $\textbf{g}$ which consists of distinct timescales representing different resistances inside the cell. These DRT curves, $\textbf{g}$, are then used as inputs to a long short-term memory (LSTM)-based neural network model for SOH estimation. We validate the model performance by testing it on ten different test sets, and achieve an average RMSPE of 1.69% across these sets.
Authors: John Hurwitz, Charles Nicholas, Edward Raff
Abstract: It is generally well understood that predictive classification and compression are intrinsically related concepts in information theory. Indeed, many deep learning methods are explained as learning a kind of compression, and that better compression leads to better performance. We interrogate this hypothesis via the Normalized Compression Distance (NCD), which explicitly relies on compression as the means of measuring similarity between sequences and thus enables nearest-neighbor classification. By turning popular large language models (LLMs) into lossless compressors, we develop a Neural NCD and compare LLMs to classic general-purpose algorithms like gzip. In doing so, we find that classification accuracy is not predictable by compression rate alone, among other empirical aberrations not predicted by current understanding. Our results imply that our intuition on what it means for a neural network to ``compress'' and what is needed for effective classification are not yet well understood.
Authors: Shirong Zheng, Shaobo Liu, Zhenhong Zhang, Dian Gu, Chunqiu Xia, Huadong Pang, Enock Mintah Ampaw
Abstract: With the advancement of global climate change and sustainable development goals, urban building energy consumption optimization and carbon emission reduction have become the focus of research. Traditional energy consumption prediction methods often lack accuracy and adaptability due to their inability to fully consider complex energy consumption patterns, especially in dealing with seasonal fluctuations and dynamic changes. This study proposes a hybrid deep learning model that combines TRIZ innovation theory with GWO, SARIMA and LSTM to improve the accuracy of building energy consumption prediction. TRIZ plays a key role in model design, providing innovative solutions to achieve an effective balance between energy efficiency, cost and comfort by systematically analyzing the contradictions in energy consumption optimization. GWO is used to optimize the parameters of the model to ensure that the model maintains high accuracy under different conditions. The SARIMA model focuses on capturing seasonal trends in the data, while the LSTM model handles short-term and long-term dependencies in the data, further improving the accuracy of the prediction. The main contribution of this research is the development of a robust model that leverages the strengths of TRIZ and advanced deep learning techniques, improving the accuracy of energy consumption predictions. Our experiments demonstrate a significant 15% reduction in prediction error compared to existing models. This innovative approach not only enhances urban energy management but also provides a new framework for optimizing energy use and reducing carbon emissions, contributing to sustainable development.
Authors: Te Li, Mengze Zhang, Yan Zhou
Abstract: Against the backdrop of increasingly severe global environmental changes, accurately predicting and meeting renewable energy demands has become a key challenge for sustainable business development. Traditional energy demand forecasting methods often struggle with complex data processing and low prediction accuracy. To address these issues, this paper introduces a novel approach that combines deep learning techniques with environmental decision support systems. The model integrates advanced deep learning techniques, including LSTM and Transformer, and PSO algorithm for parameter optimization, significantly enhancing predictive performance and practical applicability. Results show that our model achieves substantial improvements across various metrics, including a 30% reduction in MAE, a 20% decrease in MAPE, a 25% drop in RMSE, and a 35% decline in MSE. These results validate the model's effectiveness and reliability in renewable energy demand forecasting. This research provides valuable insights for applying deep learning in environmental decision support systems.
Authors: Xiaolin Lv, Liang Du, Peng Zhou, Peng Wu
Abstract: Feature selection technology is a key technology of data dimensionality reduction. Becauseof the lack of label information of collected data samples, unsupervised feature selection has attracted more attention. The universality and stability of many unsupervised feature selection algorithms are very low and greatly affected by the dataset structure. For this reason, many researchers have been keen to improve the stability of the algorithm. This paper attempts to preprocess the data set and use an interval method to approximate the data set, experimentally verifying the advantages and disadvantages of the new interval data set. This paper deals with these data sets from the global perspective and proposes a new algorithm-unsupervised feature selection algorithm based on neighborhood interval disturbance fusion(NIDF). This method can realize the joint learning of the final score of the feature and the approximate data interval. By comparing with the original unsupervised feature selection methods and several existing feature selection frameworks, the superiority of the proposed model is verified.
Authors: Wenchao Teng, Louis J. Durlofsky
Abstract: Data assimilation will be essential for the management and expansion of geological carbon storage operations. In traditional data assimilation approaches a fixed set of geological hyperparameters, such as mean and standard deviation of log-permeability, is often assumed. Such hyperparameters, however, may be highly uncertain in practical CO2 storage applications. In this study, we develop a hierarchical data assimilation framework for carbon storage that treats hyperparameters as uncertain variables characterized by hyperprior distributions. To deal with the computationally intractable likelihood function in hyperparameter estimation, we apply a likelihood-free (or simulation-based) inference algorithm, specifically sequential Monte Carlo-based approximate Bayesian computation (SMC-ABC), to draw independent posterior samples of hyperparameters given dynamic monitoring-well data. In the second step we use an ensemble smoother with multiple data assimilation (ESMDA) procedure to provide posterior realizations of grid-block permeability. To reduce computational costs, a 3D recurrent R-U-Net deep-learning surrogate model is applied for forward function evaluations. The accuracy of the surrogate model is established through comparisons to high-fidelity simulation results. A rejection sampling (RS) procedure for data assimilation is applied to provide reference posterior results. Detailed data assimilation results from SMC-ABC-ESMDA are compared to those from the reference RS method. These include marginal posterior distributions of hyperparameters, pairwise posterior samples, and history matching results for pressure and saturation at the monitoring location. Close agreement is achieved with 'converged' RS results, for two synthetic true models, in all quantities considered. Importantly, the SMC-ABC-ESMDA procedure provides speedup of 1-2 orders of magnitude relative to RS for the two cases.
Authors: Liang Du, Xin Ren, Haiying Zhang, Peng Zhou
Abstract: Multiple kernel methods less consider the intrinsic manifold structure of multiple kernel data and estimate the consensus kernel matrix with quadratic number of variables, which makes it vulnerable to the noise and outliers within multiple candidate kernels. This paper first presents the clustering method via kernelized local regression (CKLR). It captures the local structure of kernel data and employs kernel regression on the local region to predict the clustering results. Moreover, this paper further extends it to perform clustering via the multiple kernel local regression (CMKLR). We construct the kernel level local regression sparse coefficient matrix for each candidate kernel, which well characterizes the kernel level manifold structure. We then aggregate all the kernel level local regression coefficients via linear weights and generate the consensus sparse local regression coefficient, which largely reduces the number of candidate variables and becomes more robust against noises and outliers within multiple kernel data. Thus, the proposed method CMKLR avoids the above two limitations. It only contains one additional hyperparameter for tuning. Extensive experimental results show that the clustering performance of the proposed method on benchmark datasets is better than that of 10 state-of-the-art multiple kernel clustering methods.
Authors: Lei Wang, Liang Du, Peng Zhou, Peng Wu
Abstract: A symmetric nonnegative matrix factorization algorithm based on self-paced learning was proposed to improve the clustering performance of the model. It could make the model better distinguish normal samples from abnormal samples in an error-driven way. A weight variable that could measure the degree of difficulty to all samples was assigned in this method, and the variable was constrained by adopting both hard-weighting and soft-weighting strategies to ensure the rationality of the model. Cluster analysis was carried out on multiple data sets such as images and texts, and the experimental results showed the effectiveness of the proposed algorithm.
Authors: Yijie Zhang
Abstract: Bayesian inference provides a principled probabilistic framework for quantifying uncertainty by updating beliefs based on prior knowledge and observed data through Bayes' theorem. In Bayesian deep learning, neural network weights are treated as random variables with prior distributions, allowing for a probabilistic interpretation and quantification of predictive uncertainty. However, Bayesian methods lack theoretical generalization guarantees for unseen data. PAC-Bayesian analysis addresses this limitation by offering a frequentist framework to derive generalization bounds for randomized predictors, thereby certifying the reliability of Bayesian methods in machine learning. Temperature $T$, or inverse-temperature $\lambda = \frac{1}{T}$, originally from statistical mechanics in physics, naturally arises in various areas of statistical inference, including Bayesian inference and PAC-Bayesian analysis. In Bayesian inference, when $T < 1$ (``cold'' posteriors), the likelihood is up-weighted, resulting in a sharper posterior distribution. Conversely, when $T > 1$ (``warm'' posteriors), the likelihood is down-weighted, leading to a more diffuse posterior distribution. By balancing the influence of observed data and prior regularization, temperature adjustments can address issues of underfitting or overfitting in Bayesian models, bringing improved predictive performance.
Authors: Haoye Chai, Shiyuan Zhang, Xiaoqian Qi, Yong Li
Abstract: Mobile traffic forecasting allows operators to anticipate network dynamics and performance in advance, offering substantial potential for enhancing service quality and improving user experience. However, existing models are often task-oriented and are trained with tailored data, which limits their effectiveness in diverse mobile network tasks of Base Station (BS) deployment, resource allocation, energy optimization, etc. and hinders generalization across different urban environments. Foundation models have made remarkable strides across various domains of NLP and CV due to their multi-tasking adaption and zero/few-shot learning capabilities. In this paper, we propose an innovative Foundation model for Mo}bile traffic forecasting (FoMo), aiming to handle diverse forecasting tasks of short/long-term predictions and distribution generation across multiple cities to support network planning and optimization. FoMo combines diffusion models and transformers, where various spatio-temporal masks are proposed to enable FoMo to learn intrinsic features of different tasks, and a contrastive learning strategy is developed to capture the correlations between mobile traffic and urban contexts, thereby improving its transfer learning capability. Extensive experiments on 9 real-world datasets demonstrate that FoMo outperforms current models concerning diverse forecasting tasks and zero/few-shot learning, showcasing a strong universality. We further deploy the FoMo on the JiuTian optimization platform of China Mobile, where we use the predicted mobile data to formulate network planning and optimization applications, including BS deployment, resource block scheduling, and BS sleep control.
Authors: Junhao Hu, Wenrui Huang, Haoyi Wang, Weidong Wang, Tiancheng Hu, Qin Zhang, Hao Feng, Xusheng Chen, Yizhou Shan, Tao Xie
Abstract: Large Language Models (LLMs) are critical for a wide range of applications, but serving them efficiently becomes increasingly challenging as inputs become more complex. Context caching improves serving performance by exploiting inter-request dependency and reusing key-value (KV) cache across requests, thus improving time-to-first-token (TTFT). However, existing prefix-based context caching requires exact token prefix matches, limiting cache reuse in few-shot learning, multi-document QA, or retrieval-augmented generation, where prefixes may vary. In this paper, we present EPIC, an LLM serving system that introduces position-independent context caching (PIC), enabling modular KV cache reuse regardless of token chunk position (or prefix). EPIC features two key designs: AttnLink, which leverages static attention sparsity to minimize recomputation for accuracy recovery, and KVSplit, a customizable chunking method that preserves semantic coherence. Our experiments demonstrate that Epic delivers up to 8x improvements in TTFT and 7x throughput over existing systems, with negligible or no accuracy loss. By addressing the limitations of traditional caching approaches, Epic enables more scalable and efficient LLM inference.
Authors: Yara Shamshoum, Nitzan Hodos, Yuval Sieradzki, Assaf Schuster
Abstract: We introduce CompAct, a technique that reduces peak memory utilization on GPU by 25-30% for pretraining and 50% for fine-tuning of LLMs. Peak device memory is a major limiting factor in training LLMs, with various recent works aiming to reduce model memory. However most works don't target the largest component of allocated memory during training: the model's compute graph, which is stored for the backward pass. By storing low-rank, compressed activations to be used in the backward pass we greatly reduce the required memory, unlike previous methods which only reduce optimizer overheads or the number of trained parameters. Our compression uses random projection matrices, thus avoiding additional memory overheads. Comparisons with previous techniques for either pretraining or fine-tuning show that CompAct substantially improves existing compute-performance tradeoffs. We expect CompAct's savings to scale even higher for larger models.
Authors: Zhenyu Lin, Hongzheng Li, Yingxia Shao, Guanhua Ye, Yawen Li, Quanqing Xu
Abstract: Graph Contrastive Learning frameworks have demonstrated success in generating high-quality node representations. The existing research on efficient data augmentation methods and ideal pretext tasks for graph contrastive learning remains limited, resulting in suboptimal node representation in the unsupervised setting. In this paper, we introduce LAC, a graph contrastive learning framework with learnable data augmentation in an orthogonal continuous space. To capture the representative information in the graph data during augmentation, we introduce a continuous view augmenter, that applies both a masked topology augmentation module and a cross-channel feature augmentation module to adaptively augment the topological information and the feature information within an orthogonal continuous space, respectively. The orthogonal nature of continuous space ensures that the augmentation process avoids dimension collapse. To enhance the effectiveness of pretext tasks, we propose an information-theoretic principle named InfoBal and introduce corresponding pretext tasks. These tasks enable the continuous view augmenter to maintain consistency in the representative information across views while maximizing diversity between views, and allow the encoder to fully utilize the representative information in the unsupervised setting. Our experimental results show that LAC significantly outperforms the state-of-the-art frameworks.
Authors: Xiao Li, Zhuhong Li, Qiongxiu Li, Bingze Lee, Jinghao Cui, Xiaolin Hu
Abstract: Aligned Large Language Models (LLMs) have demonstrated remarkable performance across various tasks. However, LLMs remain susceptible to jailbreak adversarial attacks, where adversaries manipulate prompts to elicit malicious responses that aligned LLMs should have avoided. Identifying these vulnerabilities is crucial for understanding the inherent weaknesses of LLMs and preventing their potential misuse. One pioneering work in jailbreaking is the GCG attack, a discrete token optimization algorithm that seeks to find a suffix capable of jailbreaking aligned LLMs. Despite the success of GCG, we find it suboptimal, requiring significantly large computational costs, and the achieved jailbreaking performance is limited. In this work, we propose Faster-GCG, an efficient adversarial jailbreak method by delving deep into the design of GCG. Experiments demonstrate that Faster-GCG can surpass the original GCG with only 1/10 of the computational cost, achieving significantly higher attack success rates on various open-source aligned LLMs. In addition, We demonstrate that Faster-GCG exhibits improved attack transferability when testing on closed-sourced LLMs such as ChatGPT.
Authors: Jiangtao Kong, Jiacheng Shi, Ashley Gao, Shaohan Hu, Tianyi Zhou, Huajie Shao
Abstract: Incremental learning (IL) aims to acquire new knowledge from current tasks while retaining knowledge learned from previous tasks. Replay-based IL methods store a set of exemplars from previous tasks in a buffer and replay them when learning new tasks. However, there is usually a size-limited buffer that cannot store adequate real exemplars to retain the knowledge of previous tasks. In contrast, data distillation (DD) can reduce the exemplar buffer's size, by condensing a large real dataset into a much smaller set of more information-compact synthetic exemplars. Nevertheless, DD's performance gain on IL quickly vanishes as the number of synthetic exemplars grows. To overcome the weaknesses of real-data and synthetic-data buffers, we instead optimize a hybrid memory including both types of data. Specifically, we propose an innovative modification to DD that distills synthetic data from a sliding window of checkpoints in history (rather than checkpoints on multiple training trajectories). Conditioned on the synthetic data, we then optimize the selection of real exemplars to provide complementary improvement to the DD objective. The optimized hybrid memory combines the strengths of synthetic and real exemplars, effectively mitigating catastrophic forgetting in Class IL (CIL) when the buffer size for exemplars is limited. Notably, our method can be seamlessly integrated into most existing replay-based CIL models. Extensive experiments across multiple benchmarks demonstrate that our method significantly outperforms existing replay-based baselines.
Authors: Seyed Mohammad Ahmadi, Koorosh Aslansefat, Ruben Valcarce-Dineiro, Joshua Barnfather
Abstract: In today's world, the significance of explainable AI (XAI) is growing in robotics and point cloud applications, as the lack of transparency in decision-making can pose considerable safety risks, particularly in autonomous systems. As these technologies are integrated into real-world environments, ensuring that model decisions are interpretable and trustworthy is vital for operational reliability and safety assurance. This study explores the implementation of SMILE, a novel explainability method originally designed for deep neural networks, on point cloud-based models. SMILE builds on LIME by incorporating Empirical Cumulative Distribution Function (ECDF) statistical distances, offering enhanced robustness and interpretability, particularly when the Anderson-Darling distance is used. The approach demonstrates superior performance in terms of fidelity loss, R2 scores, and robustness across various kernel widths, perturbation numbers, and clustering configurations. Moreover, this study introduces a stability analysis for point cloud data using the Jaccard index, establishing a new benchmark and baseline for model stability in this field. The study further identifies dataset biases in the classification of the 'person' category, emphasizing the necessity for more comprehensive datasets in safety-critical applications like autonomous driving and robotics. The results underscore the potential of advanced explainability models and highlight areas for future research, including the application of alternative surrogate models and explainability techniques in point cloud data.
Authors: Xinyu Liang, Ziheng Wang, Hao Wang
Abstract: Generating synthetic residential load data that can accurately represent actual electricity consumption patterns is crucial for effective power system planning and operation. The necessity for synthetic data is underscored by the inherent challenges associated with using real-world load data, such as privacy considerations and logistical complexities in large-scale data collection. In this work, we tackle the above-mentioned challenges by developing the Ensemble Recurrent Generative Adversarial Network (ERGAN) framework to generate high-fidelity synthetic residential load data. ERGAN leverages an ensemble of recurrent Generative Adversarial Networks, augmented by a loss function that concurrently takes into account adversarial loss and differences between statistical properties. Our developed ERGAN can capture diverse load patterns across various households, thereby enhancing the realism and diversity of the synthetic data generated. Comprehensive evaluations demonstrate that our method consistently outperforms established benchmarks in the synthetic generation of residential load data across various performance metrics including diversity, similarity, and statistical measures. The findings confirm the potential of ERGAN as an effective tool for energy applications requiring synthetic yet realistic load data. We also make the generated synthetic residential load patterns publicly available.
Authors: Yingjun Du, Wenfang Sun, Cees G. M. Snoek
Abstract: Pre-trained vision-language models like CLIP have remarkably adapted to various downstream tasks. Nonetheless, their performance heavily depends on the specificity of the input text prompts, which requires skillful prompt template engineering. Instead, current approaches to prompt optimization learn the prompts through gradient descent, where the prompts are treated as adjustable parameters. However, these methods tend to lead to overfitting of the base classes seen during training and produce prompts that are no longer understandable by humans. This paper introduces a simple but interpretable prompt optimizer (IPO), that utilizes large language models (LLMs) to generate textual prompts dynamically. We introduce a Prompt Optimization Prompt that not only guides LLMs in creating effective prompts but also stores past prompts with their performance metrics, providing rich in-context information. Additionally, we incorporate a large multimodal model (LMM) to condition on visual content by generating image descriptions, which enhance the interaction between textual and visual modalities. This allows for thae creation of dataset-specific prompts that improve generalization performance, while maintaining human comprehension. Extensive testing across 11 datasets reveals that IPO not only improves the accuracy of existing gradient-descent-based prompt learning methods but also considerably enhances the interpretability of the generated prompts. By leveraging the strengths of LLMs, our approach ensures that the prompts remain human-understandable, thereby facilitating better transparency and oversight for vision-language models.
Authors: Bar Avraham, Yisroel Mirsky
Abstract: Black box attacks, where adversaries have limited knowledge of the target model, pose a significant threat to machine learning systems. Adversarial examples generated with a substitute model often suffer from limited transferability to the target model. While recent work explores ranking perturbations for improved success rates, these methods see only modest gains. We propose a novel strategy called PEAS that can boost the transferability of existing black box attacks. PEAS leverages the insight that samples which are perceptually equivalent exhibit significant variability in their adversarial transferability. Our approach first generates a set of images from an initial sample via subtle augmentations. We then evaluate the transferability of adversarial perturbations on these images using a set of substitute models. Finally, the most transferable adversarial example is selected and used for the attack. Our experiments show that PEAS can double the performance of existing attacks, achieving a 2.5x improvement in attack success rates on average over current ranking methods. We thoroughly evaluate PEAS on ImageNet and CIFAR-10, analyze hyperparameter impacts, and provide an ablation study to isolate each component's importance.
Authors: Abdul-Kazeem Shamba, Kerstin Bach, Gavin Taylor
Abstract: Understanding events in time series is an important task in a variety of contexts. However, human analysis and labeling are expensive and time-consuming. Therefore, it is advantageous to learn embeddings for moments in time series in an unsupervised way, which allows for good performance in classification or detection tasks after later minimal human labeling. In this paper, we propose dynamic contrastive learning (DynaCL), an unsupervised contrastive representation learning framework for time series that uses temporal adjacent steps to define positive pairs. DynaCL adopts N-pair loss to dynamically treat all samples in a batch as positive or negative pairs, enabling efficient training and addressing the challenges of complicated sampling of positives. We demonstrate that DynaCL embeds instances from time series into semantically meaningful clusters, which allows superior performance on downstream tasks on a variety of public time series datasets. Our findings also reveal that high scores on unsupervised clustering metrics do not guarantee that the representations are useful in downstream tasks.
Authors: Gavin Ruan, Ziqi Guo, Guang Lin
Abstract: Over 44 million Americans currently suffer from food insecurity, of whom 13 million are children. Across the United States, thousands of food banks and pantries serve as vital sources of food and other forms of aid for food insecure families. By optimizing food bank and pantry locations, food would become more accessible to families who desperately require it. In this work, we introduce a novel two-level optimization framework, which utilizes the K-Medoids clustering algorithm in conjunction with the Open-Source Routing Machine engine, to optimize food bank and pantry locations based on real road distances to houses and house blocks. Our proposed framework also has the adaptability to factor in considerations such as median household income using a pseudo-weighted K-Medoids algorithm. Testing conducted with California and Indiana household data, as well as comparisons with real food bank and pantry locations showed that interestingly, our proposed framework yields food pantry locations superior to those of real existing ones and saves significant distance for households, while there is a marginal penalty on the first level food bank to food pantry distance. Overall, we believe that the second-level benefits of this framework far outweigh any drawbacks and yield a net benefit result.
Authors: Christina Hastings Blow, Lijun Qian, Camille Gibson, Pamela Obiomon, Xishuang Dong
Abstract: AI fairness seeks to improve the transparency and explainability of AI systems by ensuring that their outcomes genuinely reflect the best interests of users. Data augmentation, which involves generating synthetic data from existing datasets, has gained significant attention as a solution to data scarcity. In particular, diffusion models have become a powerful technique for generating synthetic data, especially in fields like computer vision. This paper explores the potential of diffusion models to generate synthetic tabular data to improve AI fairness. The Tabular Denoising Diffusion Probabilistic Model (Tab-DDPM), a diffusion model adaptable to any tabular dataset and capable of handling various feature types, was utilized with different amounts of generated data for data augmentation. Additionally, reweighting samples from AIF360 was employed to further enhance AI fairness. Five traditional machine learning models-Decision Tree (DT), Gaussian Naive Bayes (GNB), K-Nearest Neighbors (KNN), Logistic Regression (LR), and Random Forest (RF)-were used to validate the proposed approach. Experimental results demonstrate that the synthetic data generated by Tab-DDPM improves fairness in binary classification.
Authors: Peng Wu, Tales Imbiriba, Pau Closas
Abstract: One of the main challenges of federated learning (FL) is handling non-independent and identically distributed (non-IID) client data, which may occur in practice due to unbalanced datasets and use of different data sources across clients. Knowledge sharing and model personalization are key strategies for addressing this issue. Clustered federated learning is a class of FL methods that groups clients that observe similarly distributed data into clusters, such that every client is typically associated with one data distribution and participates in training a model for that distribution along their cluster peers. In this paper, we present a unified Bayesian framework for clustered FL which associates clients to clusters. Then we propose several practical algorithms to handle the, otherwise growing, data associations in a way that trades off performance and computational complexity. This work provides insights on client-cluster associations and enables client knowledge sharing in new ways. The proposed framework circumvents the need for unique client-cluster associations, which is seen to increase the performance of the resulting models in a variety of experiments.
Authors: Timofei Gritsaev, Nikita Morozov, Sergey Samsonov, Daniil Tiapkin
Abstract: Generative Flow Networks (GFlowNets) are a family of generative models that learn to sample objects with probabilities proportional to a given reward function. The key concept behind GFlowNets is the use of two stochastic policies: a forward policy, which incrementally constructs compositional objects, and a backward policy, which sequentially deconstructs them. Recent results show a close relationship between GFlowNet training and entropy-regularized reinforcement learning (RL) problems with a particular reward design. However, this connection applies only in the setting of a fixed backward policy, which might be a significant limitation. As a remedy to this problem, we introduce a simple backward policy optimization algorithm that involves direct maximization of the value function in an entropy-regularized Markov Decision Process (MDP) over intermediate rewards. We provide an extensive experimental evaluation of the proposed approach across various benchmarks in combination with both RL and GFlowNet algorithms and demonstrate its faster convergence and mode discovery in complex environments.
Authors: Heshan Fernando, Han Shen, Parikshit Ram, Yi Zhou, Horst Samulowitz, Nathalie Baracaldo, Tianyi Chen
Abstract: Post-training of pre-trained LLMs, which typically consists of the supervised fine-tuning (SFT) stage and the preference learning (RLHF or DPO) stage, is crucial to effective and safe LLM applications. The widely adopted approach in post-training popular open-source LLMs is to sequentially perform SFT and RLHF/DPO. However, sequential training is sub-optimal in terms of SFT and RLHF/DPO trade-off: the LLM gradually forgets about the first stage's training when undergoing the second stage's training. We theoretically prove the sub-optimality of sequential post-training. Furthermore, we propose a practical joint post-training framework with theoretical convergence guarantees and empirically outperforms sequential post-training framework, while having similar computational cost. Our code is available at https://github.com/heshandevaka/XRIGHT.
Authors: Sanchit Sinha, Guangzhi Xiong, Aidong Zhang
Abstract: The rising need for explainable deep neural network architectures has utilized semantic concepts as explainable units. Several approaches utilizing disentangled representation learning estimate the generative factors and utilize them as concepts for explaining DNNs. However, even though the generative factors for a dataset remain fixed, concepts are not fixed entities and vary based on downstream tasks. In this paper, we propose a disentanglement mechanism utilizing a variational autoencoder (VAE) for learning mutually independent generative factors for a given dataset and subsequently learning task-specific concepts using a structural causal model (SCM). Our method assumes generative factors and concepts to form a bipartite graph, with directed causal edges from generative factors to concepts. Experiments are conducted on datasets with known generative factors: D-sprites and Shapes3D. On specific downstream tasks, our proposed method successfully learns task-specific concepts which are explained well by the causal edges from the generative factors. Lastly, separate from current causal concept discovery methods, our methodology is generalizable to an arbitrary number of concepts and flexible to any downstream tasks.
Authors: Arisrei Lim, Abhiram Maddukuri
Abstract: In recent years, reinforcement learning (RL) has gained popularity and has been applied to a wide range of tasks. One such popular domain where RL has been effective is resource management problems in systems. We look to extend work on RL for resource management problems by considering the novel domain of dynamic memory allocation management. We consider dynamic memory allocation to be a suitable domain for RL since current algorithms like first-fit, best-fit, and worst-fit can fail to adapt to changing conditions and can lead to fragmentation and suboptimal efficiency. In this paper, we present a framework in which an RL agent continuously learns from interactions with the system to improve memory management tactics. We evaluate our approach through various experiments using high-level and low-level action spaces and examine different memory allocation patterns. Our results show that RL can successfully train agents that can match and surpass traditional allocation strategies, particularly in environments characterized by adversarial request patterns. We also explore the potential of history-aware policies that leverage previous allocation requests to enhance the allocator's ability to handle complex request patterns. Overall, we find that RL offers a promising avenue for developing more adaptive and efficient memory allocation strategies, potentially overcoming limitations of hardcoded allocation algorithms.
Authors: Parsa Esmati, Amirhossein Dadashzadeh, Vahid Goodarzi, Nicolas Larrosa, Nicolo Grilli
Abstract: Current approaches using sequential networks have shown promise in estimating field variables for dynamical systems, but they are often limited by high rollout errors. The unresolved issue of rollout error accumulation results in unreliable estimations as the network predicts further into the future, with each step's error compounding and leading to an increase in inaccuracy. Here, we introduce the State-Exchange Attention (SEA) module, a novel transformer-based module enabling information exchange between encoded fields through multi-head cross-attention. The cross-field multidirectional information exchange design enables all state variables in the system to exchange information with one another, capturing physical relationships and symmetries between fields. In addition, we incorporate a ViT-like architecture to generate spatially coherent mesh embeddings, further improving the model's ability to capture spatial dependencies in the data. This enhances the model's ability to represent complex interactions between the field variables, resulting in improved rollout error accumulation. Our results show that the Transformer model integrated with the State-Exchange Attention (SEA) module outperforms competitive baseline models, including the PbGMR-GMUS Transformer-RealNVP and GMR-GMUS Transformer, with a reduction in error of 88\% and 91\%, respectively, achieving state-of-the-art performance. Furthermore, we demonstrate that the SEA module alone can reduce errors by 97\% for state variables that are highly dependent on other states of the system.
Authors: Rohan Saha, Abrar Fahim, Alona Fyshe, Alex Murphy
Abstract: For specialized domains, there is often not a wealth of data with which to train large machine learning models. In such limited data / compute settings, various methods exist aiming to $\textit{do more with less}$, such as finetuning from a pretrained model, modulating difficulty levels as data are presented to a model (curriculum learning), and considering the role of model type / size. Approaches to efficient $\textit{machine}$ learning also take inspiration from $\textit{human}$ learning by considering use cases where machine learning systems have access to approximately the same number of words experienced by a 13 year old child (100M words). We investigate the role of 3 primary variables in a limited data regime as part of the multimodal track of the BabyLM challenge. We contrast: (i) curriculum learning, (ii), pretraining (with text-only data), (iii) model type. We modulate these variables and assess them on two types of tasks: (a) multimodal (text+image), and (b) unimodal (text-only) tasks. We find that curriculum learning benefits multimodal evaluations over non-curriclum learning models, particularly when combining text-only pretraining. On text-only tasks, curriculum learning appears to help models with smaller trainable parameter counts. We suggest possible reasons based on architectural differences and training designs as to why one might observe such results.
Authors: Ange-Cl\'ement Akazan, Alexia Jolicoeur-Martineau, Ioannis Mitliagkas
Abstract: Privacy and regulatory constraints make data generation vital to advancing machine learning without relying on real-world datasets. A leading approach for tabular data generation is the Forest Flow (FF) method, which combines Flow Matching with XGBoost. Despite its good performance, FF is slow and makes errors when treating categorical variables as one-hot continuous features. It is also highly sensitive to small changes in the initial conditions of the ordinary differential equation (ODE). To overcome these limitations, we develop Heterogeneous Sequential Feature Forest Flow (HS3F). Our method generates data sequentially (feature-by-feature), reducing the dependency on noisy initial conditions through the additional information from previously generated features. Furthermore, it generates categorical variables using multinomial sampling (from an XGBoost classifier) instead of flow matching, improving generation speed. We also use a Runge-Kutta 4th order (Rg4) ODE solver for improved performance over the Euler solver used in FF. Our experiments with 25 datasets reveal that HS3F produces higher quality and more diverse synthetic data than FF, especially for categorical variables. It also generates data 21-27 times faster for datasets with $\geq20%$ categorical variables. HS3F further demonstrates enhanced robustness to affine transformation in flow ODE initial conditions compared to FF. This study not only validates the HS3F but also unveils promising new strategies to advance generative models.
Authors: Ahmed Elbakary, Chaouki Ben Issaid, Tamer ElBatt, Karim Seddik, Mehdi Bennis
Abstract: In this paper, we introduce a method for fine-tuning Large Language Models (LLMs), inspired by Multi-Task learning in a federated manner. Our approach leverages the structure of each client's model and enables a learning scheme that considers other clients' tasks and data distribution. To mitigate the extensive computational and communication overhead often associated with LLMs, we utilize a parameter-efficient fine-tuning method, specifically Low-Rank Adaptation (LoRA), reducing the number of trainable parameters. Experimental results, with different datasets and models, demonstrate the proposed method's effectiveness compared to existing frameworks for federated fine-tuning of LLMs in terms of average and local performances. The proposed scheme outperforms existing baselines by achieving lower local loss for each client while maintaining comparable global performance.
Authors: Jinda Jia, Cong Xie, Hanlin Lu, Daoce Wang, Hao Feng, Chengming Zhang, Baixi Sun, Haibin Lin, Zhi Zhang, Xin Liu, Dingwen Tao
Abstract: Recent years have witnessed a clear trend towards language models with an ever-increasing number of parameters, as well as the growing training overhead and memory usage. Distributed training, particularly through Sharded Data Parallelism (ShardedDP) which partitions optimizer states among workers, has emerged as a crucial technique to mitigate training time and memory usage. Yet, a major challenge in the scalability of ShardedDP is the intensive communication of weights and gradients. While compression techniques can alleviate this issue, they often result in worse accuracy. Driven by this limitation, we propose SDP4Bit (Toward 4Bit Communication Quantization in Sharded Data Parallelism for LLM Training), which effectively reduces the communication of weights and gradients to nearly 4 bits via two novel techniques: quantization on weight differences, and two-level gradient smooth quantization. Furthermore, SDP4Bit presents an algorithm-system co-design with runtime optimization to minimize the computation overhead of compression. In addition to the theoretical guarantees of convergence, we empirically evaluate the accuracy of SDP4Bit on the pre-training of GPT models with up to 6.7 billion parameters, and the results demonstrate a negligible impact on training loss. Furthermore, speed experiments show that SDP4Bit achieves up to 4.08$\times$ speedup in end-to-end throughput on a scale of 128 GPUs.
Authors: Saba Zerefa, Zhaolin Ren, Haitong Ma, Na Li
Abstract: In Bayesian optimization, a black-box function is maximized via the use of a surrogate model. We apply distributed Thompson sampling, using a Gaussian process as a surrogate model, to approach the multi-agent Bayesian optimization problem. In our distributed Thompson sampling implementation, each agent receives sampled points from neighbors, where the communication network is encoded in a graph; each agent utilizes a Gaussian process to model the objective function. We demonstrate a theoretical bound on Bayesian Simple Regret, where the bound depends on the size of the largest complete subgraph of the communication graph. Unlike in batch Bayesian optimization, this bound is applicable in cases where the communication graph amongst agents is constrained. When compared to sequential Thompson sampling, our bound guarantees faster convergence with respect to time as long as there is a fully connected subgraph of at least two agents. We confirm the efficacy of our algorithm with numerical simulations on traditional optimization test functions, illustrating the significance of graph connectivity on improving regret convergence.
Authors: Zhang Minghao, Song Bifeng, Yang Xiaojun, Wang Liang
Abstract: The nonlinear and unstable aerodynamic interference generated by the tandem wings of such biomimetic systems poses substantial challenges for motion control, especially under multiple random operating conditions. To address these challenges, the Concerto Reinforcement Learning Extension (CRL2E) algorithm has been developed. This plug-and-play, fully on-the-job, real-time reinforcement learning algorithm incorporates a novel Physics-Inspired Rule-Based Policy Composer Strategy with a Perturbation Module alongside a lightweight network optimized for real-time control. To validate the performance and the rationality of the module design, experiments were conducted under six challenging operating conditions, comparing seven different algorithms. The results demonstrate that the CRL2E algorithm achieves safe and stable training within the first 500 steps, improving tracking accuracy by 14 to 66 times compared to the Soft Actor-Critic, Proximal Policy Optimization, and Twin Delayed Deep Deterministic Policy Gradient algorithms. Additionally, CRL2E significantly enhances performance under various random operating conditions, with improvements in tracking accuracy ranging from 8.3% to 60.4% compared to the Concerto Reinforcement Learning (CRL) algorithm. The convergence speed of CRL2E is 36.11% to 57.64% faster than the CRL algorithm with only the Composer Perturbation and 43.52% to 65.85% faster than the CRL algorithm when both the Composer Perturbation and Time-Interleaved Capability Perturbation are introduced, especially in conditions where the standard CRL struggles to converge. Hardware tests indicate that the optimized lightweight network structure excels in weight loading and average inference time, meeting real-time control requirements.
Authors: Jean Feng, Avni Kothari, Luke Zier, Chandan Singh, Yan Shuo Tan
Abstract: Concept Bottleneck Models (CBMs) have been proposed as a compromise between white-box and black-box models, aiming to achieve interpretability without sacrificing accuracy. The standard training procedure for CBMs is to predefine a candidate set of human-interpretable concepts, extract their values from the training data, and identify a sparse subset as inputs to a transparent prediction model. However, such approaches are often hampered by the tradeoff between enumerating a sufficiently large set of concepts to include those that are truly relevant versus controlling the cost of obtaining concept extractions. This work investigates a novel approach that sidesteps these challenges: BC-LLM iteratively searches over a potentially infinite set of concepts within a Bayesian framework, in which Large Language Models (LLMs) serve as both a concept extraction mechanism and prior. BC-LLM is broadly applicable and multi-modal. Despite imperfections in LLMs, we prove that BC-LLM can provide rigorous statistical inference and uncertainty quantification. In experiments, it outperforms comparator methods including black-box models, converges more rapidly towards relevant concepts and away from spuriously correlated ones, and is more robust to out-of-distribution samples.
Authors: Zhimeng Jiang, Zirui Liu, Xiaotian Han, Qizhang Feng, Hongye Jin, Qiaoyu Tan, Kaixiong Zhou, Na Zou, Xia Hu
Abstract: Deep neural networks are ubiquitously adopted in many applications, such as computer vision, natural language processing, and graph analytics. However, well-trained neural networks can make prediction errors after deployment as the world changes. \textit{Model editing} involves updating the base model to correct prediction errors with less accessible training data and computational resources. Despite recent advances in model editors in computer vision and natural language processing, editable training in graph neural networks (GNNs) is rarely explored. The challenge with editable GNN training lies in the inherent information aggregation across neighbors, which can lead model editors to affect the predictions of other nodes unintentionally. In this paper, we first observe the gradient of cross-entropy loss for the target node and training nodes with significant inconsistency, which indicates that directly fine-tuning the base model using the loss on the target node deteriorates the performance on training nodes. Motivated by the gradient inconsistency observation, we propose a simple yet effective \underline{G}radient \underline{R}ewiring method for \underline{E}ditable graph neural network training, named \textbf{GRE}. Specifically, we first store the anchor gradient of the loss on training nodes to preserve the locality. Subsequently, we rewire the gradient of the loss on the target node to preserve performance on the training node using anchor gradient. Experiments demonstrate the effectiveness of GRE on various model architectures and graph datasets in terms of multiple editing situations. The source code is available at \url{https://github.com/zhimengj0326/Gradient_rewiring_editing}
URLs: https://github.com/zhimengj0326/Gradient_rewiring_editing
Authors: Yining Li, Peizhong Ju, Ness B. Shroff
Abstract: Multi-objective Markov Decision Processes (MDPs) are receiving increasing attention, as real-world decision-making problems often involve conflicting objectives that cannot be addressed by a single-objective MDP. The Pareto front identifies the set of policies that cannot be dominated, providing a foundation for finding optimal solutions that can efficiently adapt to various preferences. However, finding the Pareto front is a highly challenging problem. Most existing methods either (i) rely on traversing the continuous preference space, which is impractical and results in approximations that are difficult to evaluate against the true Pareto front, or (ii) focus solely on deterministic Pareto optimal policies, from which there are no known techniques to characterize the full Pareto front. Moreover, finding the structure of the Pareto front itself remains unclear even in the context of dynamic programming. This work addresses the challenge of efficiently discovering the Pareto front. By investigating the geometric structure of the Pareto front in MO-MDP, we uncover a key property: the Pareto front is on the boundary of a convex polytope whose vertices all correspond to deterministic policies, and neighboring vertices of the Pareto front differ by only one state-action pair of the deterministic policy, almost surely. This insight transforms the global comparison across all policies into a localized search among deterministic policies that differ by only one state-action pair, drastically reducing the complexity of searching for the exact Pareto front. We develop an efficient algorithm that identifies the vertices of the Pareto front by solving a single-objective MDP only once and then traversing the edges of the Pareto front, making it more efficient than existing methods. Our empirical studies demonstrate the effectiveness of our theoretical strategy in discovering the Pareto front.
Authors: Brian Cho, Dominik Meier, Kyra Gan, Nathan Kallus
Abstract: In multi-armed bandits, the tasks of reward maximization and pure exploration are often at odds with each other. The former focuses on exploiting arms with the highest means, while the latter may require constant exploration across all arms. In this work, we focus on good arm identification (GAI), a practical bandit inference objective that aims to label arms with means above a threshold as quickly as possible. We show that GAI can be efficiently solved by combining a reward-maximizing sampling algorithm with a novel nonparametric anytime-valid sequential test for labeling arm means. We first establish that our sequential test maintains error control under highly nonparametric assumptions and asymptotically achieves the minimax optimal e-power, a notion of power for anytime-valid tests. Next, by pairing regret-minimizing sampling schemes with our sequential test, we provide an approach that achieves minimax optimal stopping times for labeling arms with means above a threshold, under an error probability constraint. Our empirical results validate our approach beyond the minimax setting, reducing the expected number of samples for all stopping times by at least 50% across both synthetic and real-world settings.
Authors: Pu Zhao, Fei Sun, Xuan Shen, Pinrui Yu, Zhenglun Kong, Yanzhi Wang, Xue Lin
Abstract: Despite the superior performance, it is challenging to deploy foundation models or large language models (LLMs) due to their massive parameters and computations. While pruning is a promising technique to reduce model size and accelerate the inference, the traditional pruning techniques can hardly be applied for LLMs as they need to finetune the model on the full dataset with multiple epochs consuming massive data and hardware resources. To deal with this problem, post-training pruning methods are proposed to prune LLMs in one-shot without retraining. However, their accuracy after pruning may suffer from certain performance degradation due to the lack of retraining with massive data. To address this issue, in this paper, we first formulate the post-training problem for layer-wise LLM compression to simultaneously prune multiple weights in LLMs. Next, we provide an optimal solution for this problem and design our post-training pruning algorithm for both unstructured and semi-structured sparsity. Our extensive experiments demonstrate the superior performance of the proposed methods in comparison to SOTA baselines across various LLM families including transformer-based LLMs and Mamba-based LLMs. Code link: https://github.com/piuzha/APT
Authors: DongNyeong Heo, Heeyoul Choi
Abstract: The Transformer architecture has become widely adopted due to its demonstrated success, attributed to the attention mechanism at its core. Despite these successes, the attention mechanism of Transformers is associated with two well-known issues: rank-collapse and gradient vanishing. In this paper, we present a theoretical analysis that it is inherently difficult to address both issues simultaneously in the conventional attention mechanism. To handle these issues, we introduce a novel class of attention mechanism, referred to as generalized probabilistic attention mechanism (GPAM), and its dual-attention implementation within the Transformer architecture. Unlike conventional attention mechanisms, GPAM allows for negative attention scores while preserving a fixed total sum. We provide theoretical evidence that the proposed dual-attention GPAM (daGPAM) effectively mitigates both the rank-collapse and gradient vanishing issues which are difficult to resolve simultaneously with the conventional attention mechanisms. Furthermore, we empirically validate this theoretical evidence, demonstrating the superiority of daGPAM compared to other alternative attention mechanisms that were proposed to address the same issues. Additionally, we demonstrate the practical benefits of GPAM in natural language processing tasks, such as language modeling and neural machine translation.
Authors: Chunyuan Deng, Zhiqi Li, Roy Xie, Ruidi Chang, Hanjie Chen
Abstract: Large Language Models (LLMs) are thought to struggle with arithmetic learning due to the inherent differences between language modeling and numerical computation, but concrete evidence has been lacking. This work responds to this claim through a two-side experiment. We first investigate whether LLMs leverage partial products during arithmetic learning. We find that although LLMs can identify some partial products after learning, they fail to leverage them for arithmetic tasks, conversely. We then explore how LLMs approach arithmetic symbolically by breaking tasks into subgroups, hypothesizing that difficulties arise from subgroup complexity and selection. Our results show that when subgroup complexity is fixed, LLMs treat a collection of different arithmetic operations similarly. By analyzing position-level accuracy across different training sizes, we further observe that it follows a U-shaped pattern: LLMs quickly learn the easiest patterns at the first and last positions, while progressively learning the more difficult patterns in the middle positions. This suggests that LLMs select subgroup following an easy-to-hard paradigm during learning. Our work confirms that LLMs are pure symbolic learners in arithmetic tasks and underscores the importance of understanding them deeply through subgroup-level quantification.
Authors: Kishor Kumar Bhaumik, Minha Kim, Fahim Faisal Niloy, Amin Ahsan Ali, Simon S. Woo
Abstract: Traffic forecasting in Intelligent Transportation Systems (ITS) is vital for intelligent traffic prediction. Yet, ITS often relies on data from traffic sensors or vehicle devices, where certain cities might not have all those smart devices or enabling infrastructures. Also, recent studies have employed meta-learning to generalize spatial-temporal traffic networks, utilizing data from multiple cities for effective traffic forecasting for data-scarce target cities. However, collecting data from multiple cities can be costly and time-consuming. To tackle this challenge, we introduce Single Source Meta-Transfer Learning (SSMT) which relies only on a single source city for traffic prediction. Our method harnesses this transferred knowledge to enable few-shot traffic forecasting, particularly when the target city possesses limited data. Specifically, we use memory-augmented attention to store the heterogeneous spatial knowledge from the source city and selectively recall them for the data-scarce target city. We extend the idea of sinusoidal positional encoding to establish meta-learning tasks by leveraging diverse temporal traffic patterns from the source city. Moreover, to capture a more generalized representation of the positions we introduced a meta-positional encoding that learns the most optimal representation of the temporal pattern across all the tasks. We experiment on five real-world benchmark datasets to demonstrate that our method outperforms several existing methods in time series traffic prediction.
Authors: Amira Hijazi, Osman Ozaltin, Reha Uzsoy
Abstract: We present a neural network-enhanced column generation (CG) approach for a parallel machine scheduling problem. The proposed approach utilizes an encoder-decoder attention model, namely the transformer and pointer architectures, to develop job sequences with negative reduced cost and thus generate columns to add to the master problem. By training the neural network offline and using it in inference mode to predict negative reduced costs columns, we achieve significant computational time savings compared to dynamic programming (DP). Since the exact DP procedure is used to verify that no further columns with negative reduced cost can be identified at termination, the optimality guarantee of the original CG procedure is preserved. For small to medium-sized instances, our approach achieves an average 45% reduction in computation time compared to solving the subproblems with DP. Furthermore, the model generalizes not only to unseen, larger problem instances from the same probability distribution but also to instances from different probability distributions than those presented at training time. For large-sized instances, the proposed approach achieves an 80% improvement in the objective value in under 500 seconds, demonstrating both its scalability and efficiency.
Authors: Mudit Gaur, Amrit Singh Bedi, Raghu Pasupathy, Vaneet Aggarwal
Abstract: The importance of Reinforcement Learning from Human Feedback (RLHF) in aligning large language models (LLMs) with human values cannot be overstated. RLHF is a three-stage process that includes supervised fine-tuning (SFT), reward learning, and policy learning. Although there are several offline and online approaches to aligning LLMs, they often suffer from distribution shift issues. These issues arise from the inability to accurately capture the distributional interdependence between the reward learning and policy learning stages. Consequently, this has led to various approximated approaches, but the theoretical insights and motivations remain largely limited to tabular settings, which do not hold in practice. This gap between theoretical insights and practical implementations is critical. It is challenging to address this gap as it requires analyzing the performance of AI alignment algorithms in neural network-parameterized settings. Although bi-level formulations have shown promise in addressing distribution shift issues, they suffer from the hyper-gradient problem, and current approaches lack efficient algorithms to solve this. In this work, we tackle these challenges employing the bi-level formulation laid out in Kwon et al. (2024) along with the assumption \emph{Weak Gradient Domination} to demonstrate convergence in an RLHF setup, obtaining a sample complexity of $\epsilon^{-\frac{7}{2}}$ . Our key contributions are twofold: (i) We propose a bi-level formulation for AI alignment in parameterized settings and introduce a first-order approach to solve this problem. (ii) We analyze the theoretical convergence rates of the proposed algorithm and derive state-of-the-art bounds. To the best of our knowledge, this is the first work to establish convergence rate bounds and global optimality for the RLHF framework in neural network-parameterized settings.
Authors: Shicheng Liu, Minghui Zhu
Abstract: Inverse reinforcement learning (IRL) aims to learn a reward function and a corresponding policy that best fit the demonstrated trajectories of an expert. However, current IRL works cannot learn incrementally from an ongoing trajectory because they have to wait to collect at least one complete trajectory to learn. To bridge the gap, this paper considers the problem of learning a reward function and a corresponding policy while observing the initial state-action pair of an ongoing trajectory and keeping updating the learned reward and policy when new state-action pairs of the ongoing trajectory are observed. We formulate this problem as an online bi-level optimization problem where the upper level dynamically adjusts the learned reward according to the newly observed state-action pairs with the help of a meta-regularization term, and the lower level learns the corresponding policy. We propose a novel algorithm to solve this problem and guarantee that the algorithm achieves sub-linear local regret $O(\sqrt{T}+\log T+\sqrt{T}\log T)$. If the reward function is linear, we prove that the proposed algorithm achieves sub-linear regret $O(\log T)$. Experiments are used to validate the proposed algorithm.
Authors: Anh Bui, Long Vuong, Khanh Doan, Trung Le, Paul Montague, Tamas Abraham, Dinh Phung
Abstract: Diffusion models excel at generating visually striking content from text but can inadvertently produce undesirable or harmful content when trained on unfiltered internet data. A practical solution is to selectively removing target concepts from the model, but this may impact the remaining concepts. Prior approaches have tried to balance this by introducing a loss term to preserve neutral content or a regularization term to minimize changes in the model parameters, yet resolving this trade-off remains challenging. In this work, we propose to identify and preserving concepts most affected by parameter changes, termed as \textit{adversarial concepts}. This approach ensures stable erasure with minimal impact on the other concepts. We demonstrate the effectiveness of our method using the Stable Diffusion model, showing that it outperforms state-of-the-art erasure methods in eliminating unwanted content while maintaining the integrity of other unrelated elements. Our code is available at \url{https://github.com/tuananhbui89/Erasing-Adversarial-Preservation}.
URLs: https://github.com/tuananhbui89/Erasing-Adversarial-Preservation
Authors: Haobin Li, Peng Hu, Qianjun Zhang, Xi Peng, Xiting Liu, Mouxing Yang
Abstract: The success of most existing cross-modal retrieval methods heavily relies on the assumption that the given queries follow the same distribution of the source domain. However, such an assumption is easily violated in real-world scenarios due to the complexity and diversity of queries, thus leading to the query shift problem. Specifically, query shift refers to the online query stream originating from the domain that follows a different distribution with the source one. In this paper, we observe that query shift would not only diminish the uniformity (namely, within-modality scatter) of the query modality but also amplify the gap between query and gallery modalities. Based on the observations, we propose a novel method dubbed Test-time adaptation for Cross-modal Retrieval (TCR). In brief, TCR employs a novel module to refine the query predictions (namely, retrieval results of the query) and a joint objective to prevent query shift from disturbing the common space, thus achieving online adaptation for the cross-modal retrieval models with query shift. Expensive experiments demonstrate the effectiveness of the proposed TCR against query shift. The code will be released upon acceptance.
Authors: Anjiang Wei, Allen Nie, Thiago S. F. X. Teixeira, Rohan Yadav, Wonchan Lee, Ke Wang, Alex Aiken
Abstract: Mapping computations to processors and assigning data to memory are critical for maximizing performance in parallel programming. These mapping decisions are managed through the development of specialized low-level system code, called mappers, crafted by performance engineers. Each mapper is tailored to a specific application and optimized for the underlying machine architecture, a process that requires days of refinement and tuning from an expert. Despite advances in system research, automating mapper generation remains a challenge due to the complexity of making millions of decisions to find the optimal solution and generate the solution as code. We introduce an approach that leverages recent advances in LLM-based optimizers for mapper design. In under ten minutes, our method automatically discovers mappers that surpass human expert designs in scientific applications by up to 1.34X speedup. For parallel matrix multiplication algorithms, our mapper achieves up to 1.31X of the expert-designed solution. To achieve this, we simplify the complexity of low-level code generation by introducing a domain-specific language (DSL) that abstracts the low-level system programming details and defines a structured search space for LLMs to explore. To maximize the application performance, we use an LLM optimizer to improve an agentic system that generates the mapper code. As a result, this approach significantly reduces the workload for performance engineers while achieving substantial performance gains across diverse applications. Finally, our results demonstrate the effectiveness of LLM-based optimization in system design and suggest its potential for addressing other complex system challenges.
Authors: Aleksandar Armacki, Shuhua Yu, Dragana Bajovic, Dusan Jakovetic, Soummya Kar
Abstract: We study large deviations and mean-squared error (MSE) guarantees of a general framework of nonlinear stochastic gradient methods in the online setting, in the presence of heavy-tailed noise. Unlike existing works that rely on the closed form of a nonlinearity (typically clipping), our framework treats the nonlinearity in a black-box manner, allowing us to provide unified guarantees for a broad class of bounded nonlinearities, including many popular ones, like sign, quantization, normalization, as well as component-wise and joint clipping. We provide several strong results for a broad range of step-sizes in the presence of heavy-tailed noise with symmetric probability density function, positive in a neighbourhood of zero and potentially unbounded moments. In particular, for non-convex costs we provide a large deviation upper bound for the minimum norm-squared of gradients, showing an asymptotic tail decay on an exponential scale, at a rate $\sqrt{t} / \log(t)$. We establish the accompanying rate function, showing an explicit dependence on the choice of step-size, nonlinearity, noise and problem parameters. Next, for non-convex costs and the minimum norm-squared of gradients, we derive the optimal MSE rate $\widetilde{\mathcal{O}}(t^{-1/2})$. Moreover, for strongly convex costs and the last iterate, we provide an MSE rate that can be made arbitrarily close to the optimal rate $\mathcal{O}(t^{-1})$, improving on the state-of-the-art results in the presence of heavy-tailed noise. Finally, we establish almost sure convergence of the minimum norm-squared of gradients, providing an explicit rate, which can be made arbitrarily close to $o(t^{-1/4})$.
Authors: Jun Kato, Airi Mita, Keita Gobara, Akihiro Inokuchi
Abstract: Graphs are useful for representing various realworld objects. However, graph neural networks (GNNs) tend to suffer from over-smoothing, where the representations of nodes of different classes become similar as the number of layers increases, leading to performance degradation. A method that does not require protracted tuning of the number of layers is needed to effectively construct a graph attention network (GAT), a type of GNN. Therefore, we introduce a method called "DeepGAT" for predicting the class to which nodes belong in a deep GAT. It avoids over-smoothing in a GAT by ensuring that nodes in different classes are not similar at each layer. Using DeepGAT to predict class labels, a 15-layer network is constructed without the need to tune the number of layers. DeepGAT prevented over-smoothing and achieved a 15-layer GAT with similar performance to a 2-layer GAT, as indicated by the similar attention coefficients. DeepGAT enables the training of a large network to acquire similar attention coefficients to a network with few layers. It avoids the over-smoothing problem and obviates the need to tune the number of layers, thus saving time and enhancing GNN performance.
Authors: Debo Cheng, Ziqi Xu, Jiuyong Li, Lin Liu, Kui Yu, Thuc Duy Le, Jixue Liu
Abstract: Intervention intuition is often used in model explanation where the intervention effect of a feature on the outcome is quantified by the difference of a model prediction when the feature value is changed from the current value to the baseline value. Such a model intervention effect of a feature is inherently association. In this paper, we will study the conditions when an intuitive model intervention effect has a causal interpretation, i.e., when it indicates whether a feature is a direct cause of the outcome. This work links the model intervention effect to the causal interpretation of a model. Such an interpretation capability is important since it indicates whether a machine learning model is trustworthy to domain experts. The conditions also reveal the limitations of using a model intervention effect for causal interpretation in an environment with unobserved features. Experiments on semi-synthetic datasets have been conducted to validate theorems and show the potential for using the model intervention effect for model interpretation.
Authors: Jin Zhou (Jiaxun), Hanmei Yang (Jiaxun), Steven (Jiaxun), Tang, Mingcan Xiang, Hui Guan, Tongping Liu
Abstract: Fine-tuning with Reinforcement Learning with Human Feedback (RLHF) is essential for aligning large language models (LLMs). However, RLHF often encounters significant memory challenges. This study is the first to examine memory usage in the RLHF context, exploring various memory management strategies and unveiling the reasons behind excessive memory consumption. Additionally, we introduce a simple yet effective approach that substantially reduces the memory required for RLHF fine-tuning.
Authors: Khurram Yamin, Vibhhu Sharma, Ed Kennedy, Bryan Wilder
Abstract: Many applications of causal inference require using treatment effects estimated on a study population to make decisions in a separate target population. We consider the challenging setting where there are covariates that are observed in the target population that were not seen in the original study. Our goal is to estimate the tightest possible bounds on heterogeneous treatment effects conditioned on such newly observed covariates. We introduce a novel partial identification strategy based on ideas from ecological inference; the main idea is that estimates of conditional treatment effects for the full covariate set must marginalize correctly when restricted to only the covariates observed in both populations. Furthermore, we introduce a bias-corrected estimator for these bounds and prove that it enjoys fast convergence rates and statistical guarantees (e.g., asymptotic normality). Experimental results on both real and synthetic data demonstrate that our framework can produce bounds that are much tighter than would otherwise be possible.
Authors: Daehwan Kim, Haejun Chung, Ikbeom Jang
Abstract: Recent studies have shown that deep neural networks are not well-calibrated and produce over-confident predictions. The miscalibration issue primarily stems from the minimization of cross-entropy, which aims to align predicted softmax probabilities with one-hot labels. In ordinal regression tasks, this problem is compounded by an additional challenge: the expectation that softmax probabilities should exhibit unimodal distribution is not met with cross-entropy. Rather, the ordinal regression literature has focused on unimodality and overlooked calibration. To address these issues, we propose a novel loss function that introduces order-aware calibration, ensuring that prediction confidence adheres to ordinal relationships between classes. It incorporates soft ordinal encoding and label-smoothing-based regularization to enforce both calibration and unimodality. Extensive experiments across three popular ordinal regression benchmarks demonstrate that our approach achieves state-of-the-art calibration without compromising accuracy.
Authors: Nishant S. Gaikwad, Lucas Heublein, Nisha L. Raichur, Tobias Feigl, Christopher Mutschler, Felix Ott
Abstract: Federated learning (FL) enables multiple devices to collaboratively train a global model while maintaining data on local servers. Each device trains the model on its local server and shares only the model updates (i.e., gradient weights) during the aggregation step. A significant challenge in FL is managing the feature distribution of novel, unbalanced data across devices. In this paper, we propose an FL approach using few-shot learning and aggregation of the model weights on a global server. We introduce a dynamic early stopping method to balance out-of-distribution classes based on representation learning, specifically utilizing the maximum mean discrepancy of feature embeddings between local and global models. An exemplary application of FL is orchestrating machine learning models along highways for interference classification based on snapshots from global navigation satellite system (GNSS) receivers. Extensive experiments on four GNSS datasets from two real-world highways and controlled environments demonstrate that our FL method surpasses state-of-the-art techniques in adapting to both novel interference classes and multipath scenarios.
Authors: Sarwan Ali, Prakash Chourasia, Haris Mansoor, Bipin koirala, Murray Patterson
Abstract: The t-Distributed Stochastic Neighbor Embedding (t-SNE) has emerged as a popular dimensionality reduction technique for visualizing high-dimensional data. It computes pairwise similarities between data points by default using an RBF kernel and random initialization (in low-dimensional space), which successfully captures the overall structure but may struggle to preserve the local structure efficiently. This research proposes a novel approach called the Modified Isolation Kernel (MIK) as an alternative to the Gaussian kernel, which is built upon the concept of the Isolation Kernel. MIK uses adaptive density estimation to capture local structures more accurately and integrates robustness measures. It also assigns higher similarity values to nearby points and lower values to distant points. Comparative research using the normal Gaussian kernel, the isolation kernel, and several initialization techniques, including random, PCA, and random walk initializations, are used to assess the proposed approach (MIK). Additionally, we compare the computational efficiency of all $3$ kernels with $3$ different initialization methods. Our experimental results demonstrate several advantages of the proposed kernel (MIK) and initialization method selection. It exhibits improved preservation of the local and global structure and enables better visualization of clusters and subclusters in the embedded space. These findings contribute to advancing dimensionality reduction techniques and provide researchers and practitioners with an effective tool for data exploration, visualization, and analysis in various domains.
Authors: Jifeng Hu, Sili Huang, Li Shen, Zhejian Yang, Shengchao Hu, Shisong Tang, Hechang Chen, Yi Chang, Dacheng Tao, Lichao Sun
Abstract: Continual offline reinforcement learning (CORL) has shown impressive ability in diffusion-based lifelong learning systems by modeling the joint distributions of trajectories. However, most research only focuses on limited continual task settings where the tasks have the same observation and action space, which deviates from the realistic demands of training agents in various environments. In view of this, we propose Vector-Quantized Continual Diffuser, named VQ-CD, to break the barrier of different spaces between various tasks. Specifically, our method contains two complementary sections, where the quantization spaces alignment provides a unified basis for the selective weights activation. In the quantized spaces alignment, we leverage vector quantization to align the different state and action spaces of various tasks, facilitating continual training in the same space. Then, we propose to leverage a unified diffusion model attached by the inverse dynamic model to master all tasks by selectively activating different weights according to the task-related sparse masks. Finally, we conduct extensive experiments on 15 continual learning (CL) tasks, including conventional CL task settings (identical state and action spaces) and general CL task settings (various state and action spaces). Compared with 16 baselines, our method reaches the SOTA performance.
Authors: Ankur Kumar
Abstract: KV cache compression methods have mainly relied on scalar quantization techniques to reduce the memory requirements during decoding. In this work, we apply residual vector quantization, which has been widely used for high fidelity audio compression, to compress KV cache in large language models (LLM). We adapt the standard recipe with minimal changes to compress the output of any key or value projection matrix in a pretrained LLM: we scale the vector by its standard deviation, divide channels into groups and then quantize each group with the same residual vector quantizer. We learn the codebook using exponential moving average and there are no other learnable parameters including the input and output projections normally used in a vector quantization set up. We find that a residual depth of 8 recovers most of the performance of the unquantized model. We also find that grouping non-contiguous channels together works better than grouping contiguous channels for compressing key matrix and the method further benefits from a light weight finetuning of LLM together with the quantization. Overall, the proposed technique is competitive with existing quantization methods while being much simpler and results in 5.5x compression compared to half precision.
Authors: Shutong Chen, Yang Li
Abstract: Estimating an individual's potential response to continuously varied treatments is crucial for addressing causal questions across diverse domains, from healthcare to social sciences. However, existing methods are limited either to estimating causal effects of binary treatments, or scenarios where all confounding variables are measurable. In this work, we present ContiVAE, a novel framework for estimating causal effects of continuous treatments, measured by individual dose-response curves, considering the presence of unobserved confounders using observational data. Leveraging a variational auto-encoder with a Tilted Gaussian prior distribution, ContiVAE models the hidden confounders as latent variables, and is able to predict the potential outcome of any treatment level for each individual while effectively capture the heterogeneity among individuals. Experiments on semi-synthetic datasets show that ContiVAE outperforms existing methods by up to 62%, demonstrating its robustness and flexibility. Application on a real-world dataset illustrates its practical utility.
Authors: Imanol Echeverria, Maialen Murua, Roberto Santana
Abstract: Recent advances in deep learning have shown significant potential for solving combinatorial optimization problems in real-time. Unlike traditional methods, deep learning can generate high-quality solutions efficiently, which is crucial for applications like routing and scheduling. However, existing approaches like deep reinforcement learning (RL) and behavioral cloning have notable limitations, with deep RL suffering from slow learning and behavioral cloning relying solely on expert actions, which can lead to generalization issues and neglect of the optimization objective. This paper introduces a novel offline RL method designed for combinatorial optimization problems with complex constraints, where the state is represented as a heterogeneous graph and the action space is variable. Our approach encodes actions in edge attributes and balances expected rewards with the imitation of expert solutions. We demonstrate the effectiveness of this method on job-shop scheduling and flexible job-shop scheduling benchmarks, achieving superior performance compared to state-of-the-art techniques.
Authors: Xinyu Yuan, Yan Qiao, Pei Zhao, Rongyao Hu, Benchu Zhang
Abstract: The traffic matrix estimation (TME) problem has been widely researched for decades of years. Recent progresses in deep generative models offer new opportunities to tackle TME problems in a more advanced way. In this paper, we leverage the powerful ability of denoising diffusion probabilistic models (DDPMs) on distribution learning, and for the first time adopt DDPM to address the TME problem. To ensure a good performance of DDPM on learning the distributions of TMs, we design a preprocessing module to reduce the dimensions of TMs while keeping the data variety of each OD flow. To improve the estimation accuracy, we parameterize the noise factors in DDPM and transform the TME problem into a gradient-descent optimization problem. Finally, we compared our method with the state-of-the-art TME methods using two real-world TM datasets, the experimental results strongly demonstrate the superiority of our method on both TM synthesis and TM estimation.
Authors: Shpresim Sadiku, Moritz Wagner, Sai Ganesh Nagarajan, Sebastian Pokutta
Abstract: We study the problem of finding optimal sparse, manifold-aligned counterfactual explanations for classifiers. Canonically, this can be formulated as an optimization problem with multiple non-convex components, including classifier loss functions and manifold alignment (or \emph{plausibility}) metrics. The added complexity of enforcing \emph{sparsity}, or shorter explanations, complicates the problem further. Existing methods often focus on specific models and plausibility measures, relying on convex $\ell_1$ regularizers to enforce sparsity. In this paper, we tackle the canonical formulation using the accelerated proximal gradient (APG) method, a simple yet efficient first-order procedure capable of handling smooth non-convex objectives and non-smooth $\ell_p$ (where $0 \leq p < 1$) regularizers. This enables our approach to seamlessly incorporate various classifiers and plausibility measures while producing sparser solutions. Our algorithm only requires differentiable data-manifold regularizers and supports box constraints for bounded feature ranges, ensuring the generated counterfactuals remain \emph{actionable}. Finally, experiments on real-world datasets demonstrate that our approach effectively produces sparse, manifold-aligned counterfactual explanations while maintaining proximity to the factual data and computational efficiency.
Authors: Mohammad Hasan Ahmadilivani, Jaan Raik, Masoud Daneshtalab, Maksim Jenihhin
Abstract: Growing exploitation of Machine Learning (ML) in safety-critical applications necessitates rigorous safety analysis. Hardware reliability assessment is a major concern with respect to measuring the level of safety. Quantifying the reliability of emerging ML models, including Deep Neural Networks (DNNs), is highly complex due to their enormous size in terms of the number of parameters and computations. Conventionally, Fault Injection (FI) is applied to perform a reliability measurement. However, performing FI on modern-day DNNs is prohibitively time-consuming if an acceptable confidence level is to be achieved. In order to speed up FI for large DNNs, statistical FI has been proposed. However, the run-time for the large DNN models is still considerably long. In this work, we introduce DeepVigor+, a scalable, fast and accurate semi-analytical method as an efficient alternative for reliability measurement in DNNs. DeepVigor+ implements a fault propagation analysis model and attempts to acquire Vulnerability Factors (VFs) as reliability metrics in an optimal way. The results indicate that DeepVigor+ obtains VFs for DNN models with an error less than 1\% and 14.9 up to 26.9 times fewer simulations than the best-known state-of-the-art statistical FI enabling an accurate reliability analysis for emerging DNNs within a few minutes.
Authors: Renyuan Li, Zhehui Chen, Guanyi Wang
Abstract: Multi-Output Regression (MOR) has been widely used in scientific data analysis for decision-making. Unlike traditional regression models, MOR aims to simultaneously predict multiple real-valued outputs given an input. However, the increasing dimensionality of the outputs poses significant challenges regarding interpretability and computational scalability for modern MOR applications. As a first step to address these challenges, this paper proposes a Sparse \& High-dimensional-Output REgression (SHORE) model by incorporating additional sparsity requirements to resolve the output interpretability, and then designs a computationally efficient two-stage optimization framework capable of solving SHORE with provable accuracy via compression on outputs. Theoretically, we show that the proposed framework is computationally scalable while maintaining the same order of training loss and prediction loss before-and-after compression under arbitrary or relatively weak sample set conditions. Empirically, numerical results further validate the theoretical findings, showcasing the efficiency and accuracy of the proposed framework.
Authors: Thomas George, Pierre Nodet, Alexis Bondu, Vincent Lemaire
Abstract: Mislabeled examples are ubiquitous in real-world machine learning datasets, advocating the development of techniques for automatic detection. We show that most mislabeled detection methods can be viewed as probing trained machine learning models using a few core principles. We formalize a modular framework that encompasses these methods, parameterized by only 4 building blocks, as well as a Python library that demonstrates that these principles can actually be implemented. The focus is on classifier-agnostic concepts, with an emphasis on adapting methods developed for deep learning models to non-deep classifiers for tabular data. We benchmark existing methods on (artificial) Completely At Random (NCAR) as well as (realistic) Not At Random (NNAR) labeling noise from a variety of tasks with imperfect labeling rules. This benchmark provides new insights as well as limitations of existing methods in this setup.
Authors: Marcin Sendera, Amin Sorkhei, Tomasz Ku\'smierczyk
Abstract: Function-space priors in Bayesian Neural Networks provide a more intuitive approach to embedding beliefs directly into the model's output, thereby enhancing regularization, uncertainty quantification, and risk-aware decision-making. However, imposing function-space priors on BNNs is challenging. We address this task through optimization techniques that explore how trainable activations can accommodate complex priors and match intricate target function distributions. We discuss critical learning challenges, including identifiability, loss construction, and symmetries that arise in this context. Furthermore, we enable evidence maximization to facilitate model selection by conditioning the functional priors on additional hyperparameters. Our empirical findings demonstrate that even BNNs with a single wide hidden layer, when equipped with these adaptive trainable activations and conditioning strategies, can effectively achieve high-fidelity function-space priors, providing a robust and flexible framework for enhancing Bayesian neural network performance.
Authors: Hanseul Cho, Jaeyoung Cha, Srinadh Bhojanapalli, Chulhee Yun
Abstract: Transformers often struggle with length generalization, meaning they fail to generalize to sequences longer than those encountered during training. While arithmetic tasks are commonly used to study length generalization, certain tasks are considered notoriously difficult, e.g., multi-operand addition (requiring generalization over both the number of operands and their lengths) and multiplication (requiring generalization over both operand lengths). In this work, we achieve approximately 2-3x length generalization on both tasks, which is the first such achievement in arithmetic Transformers. We design task-specific scratchpads enabling the model to focus on a fixed number of tokens per each next-token prediction step, and apply multi-level versions of Position Coupling (Cho et al., 2024; McLeish et al., 2024) to let Transformers know the right position to attend to. On the theory side, we prove that a 1-layer Transformer using our method can solve multi-operand addition, up to operand length and operand count that are exponential in embedding dimension.
Authors: Anna Sepliarskaia, Sophie Langer, Johannes Schmidt-Hieber
Abstract: We study the generalization capabilities of Group Convolutional Neural Networks (GCNNs) with ReLU activation function by deriving upper and lower bounds for their Vapnik-Chervonenkis (VC) dimension. Specifically, we analyze how factors such as the number of layers, weights, and input dimension affect the VC dimension. We further compare the derived bounds to those known for other types of neural networks. Our findings extend previous results on the VC dimension of continuous GCNNs with two layers, thereby providing new insights into the generalization properties of GCNNs, particularly regarding the dependence on the input resolution of the data.
Authors: Camiel Oerlemans, Bram Grooten, Michiel Braat, Alaa Alassi, Emilia Silvas, Decebal Constantin Mocanu
Abstract: Predicting the behavior of road users accurately is crucial to enable the safe operation of autonomous vehicles in urban or densely populated areas. Therefore, there has been a growing interest in time series motion prediction research, leading to significant advancements in state-of-the-art techniques in recent years. However, the potential of using LiDAR data to capture more detailed local features, such as a person's gaze or posture, remains largely unexplored. To address this, we develop a novel multimodal approach for motion prediction based on the PointNet foundation model architecture, incorporating local LiDAR features. Evaluation on the Waymo Open Dataset shows a performance improvement of 6.20% and 1.58% in minADE and mAP respectively, when integrated and compared with the previous state-of-the-art MTR. We open-source the code of our LiMTR model.
Authors: D. Y. C. Wang, Lars Arne Jordanger, Jerry Chun-Wei Lin
Abstract: Customer churn, particularly in the telecommunications sector, influences both costs and profits. As the explainability of models becomes increasingly important, this study emphasizes not only the explainability of customer churn through machine learning models, but also the importance of identifying multivariate patterns and setting soft bounds for intuitive interpretation. The main objective is to use a machine learning model and fuzzy-set theory with top-\textit{k} HUIM to identify highly associated patterns of customer churn with intuitive identification, referred to as Highly Associated Fuzzy Churn Patterns (HAFCP). Moreover, this method aids in uncovering association rules among multiple features across low, medium, and high distributions. Such discoveries are instrumental in enhancing the explainability of findings. Experiments show that when the top-5 HAFCPs are included in five datasets, a mixture of performance results is observed, with some showing notable improvements. It becomes clear that high importance features enhance explanatory power through their distribution and patterns associated with other features. As a result, the study introduces an innovative approach that improves the explainability and effectiveness of customer churn prediction models.
Authors: Shikhar Vashistha, Neetesh Kumar
Abstract: Traditional graph neural networks (GNNs) lack scalability and lose individual node characteristics due to over-smoothing, especially in the case of deeper networks. This results in sub-optimal feature representation, affecting the model's performance on tasks involving dynamically changing graphs. To address this issue, we present Graph Selective States Focused Attention Networks (GSANs) based neural network architecture for graph-structured data. The GSAN is enabled by multi-head masked self-attention (MHMSA) and selective state space modeling (S3M) layers to overcome the limitations of GNNs. In GSAN, the MHMSA allows GSAN to dynamically emphasize crucial node connections, particularly in evolving graph environments. The S3M layer enables the network to adjust dynamically in changing node states and improving predictions of node behavior in varying contexts without needing primary knowledge of the graph structure. Furthermore, the S3M layer enhances the generalization of unseen structures and interprets how node states influence link importance. With this, GSAN effectively outperforms inductive and transductive tasks and overcomes the issues that traditional GNNs experience. To analyze the performance behavior of GSAN, a set of state-of-the-art comparative experiments are conducted on graphs benchmark datasets, including $Cora$, $Citeseer$, $Pubmed$ network citation, and $protein-protein-interaction$ datasets, as an outcome, GSAN improved the classification accuracy by $1.56\%$, $8.94\%$, $0.37\%$, and $1.54\%$ on $F1-score$ respectively.
Authors: Aleksandra I. Nowak, Otniel-Bogdan Mercea, Anurag Arnab, Jonas Pfeiffer, Yann Dauphin, Utku Evci
Abstract: Parameter-efficient transfer learning (PETL) aims to adapt pre-trained models to new downstream tasks while minimizing the number of fine-tuned parameters. Adapters, a popular approach in PETL, inject additional capacity into existing networks by incorporating low-rank projections, achieving performance comparable to full fine-tuning with significantly fewer parameters. This paper investigates the relationship between the placement of an adapter and its performance. We observe that adapter location within a network significantly impacts its effectiveness, and that the optimal placement is task-dependent. To exploit this observation, we introduce an extended search space of adapter connections, including long-range and recurrent adapters. We demonstrate that even randomly selected adapter placements from this expanded space yield improved results, and that high-performing placements often correlate with high gradient rank. Our findings reveal that a small number of strategically placed adapters can match or exceed the performance of the common baseline of adding adapters in every block, opening a new avenue for research into optimal adapter placement strategies.
Authors: Xin Ma, Yang Liu, Jingjing Liu, Xiaoxu Ma
Abstract: Large language models (LLMs), although having revolutionized many fields, still suffer from the challenging extrapolation problem, where the inference ability of LLMs sharply declines beyond their max training lengths. In this work, we conduct a theoretical analysis to better understand why No Position Encoding (NoPE) fails outside its effective range, as well as examining the power of Position Encoding (PE) in this context. Our findings reveal that with meticulous weave position, PE can indeed be extended beyond effective range. Our theorems establish that LLMs equipped with weave PE can achieve improved extrapolation performance without additional cost. Furthermore, we introduce a novel weave PE method, Mesa-Extrapolation, which utilizes a chunk-based triangular attention matrix and applies Stair PE to manage the final chunk. This method not only retains competitive performance but also offers substantial benefits such as significantly reduced memory demand and faster inference speed. Extensive experiments validate the effectiveness of Mesa-Extrapolation, demonstrating its potential as a scalable solution to enhancing LLMs applicative reach.
Authors: Olivier Graffeuille, Yun Sing Koh, Joerg Wicker, Moritz Lehmann
Abstract: Knowledge transfer in multi-task learning is typically viewed as a dichotomy; positive transfer, which improves the performance of all tasks, or negative transfer, which hinders the performance of all tasks. In this paper, we investigate the understudied problem of asymmetric task relationships, where knowledge transfer aids the learning of certain tasks while hindering the learning of others. We propose an optimisation strategy that includes additional cloned tasks named self-auxiliaries into the learning process to flexibly transfer knowledge between tasks asymmetrically. Our method can exploit asymmetric task relationships, benefiting from the positive transfer component while avoiding the negative transfer component. We demonstrate that asymmetric knowledge transfer provides substantial improvements in performance compared to existing multi-task optimisation strategies on benchmark computer vision problems.
Authors: Woosung Koh, Wonbeen Oh, Siyeol Kim, Suhin Shin, Hyeongjin Kim, Jaein Jang, Junghyun Lee, Se-Young Yun
Abstract: Multi-agent reinforcement learning has demonstrated significant potential in addressing complex cooperative tasks across various real-world applications. However, existing MARL approaches often rely on the restrictive assumption that the number of entities (e.g., agents, obstacles) remains constant between training and inference. This overlooks scenarios where entities are dynamically removed or added during the inference trajectory -- a common occurrence in real-world environments like search and rescue missions and dynamic combat situations. In this paper, we tackle the challenge of intra-trajectory dynamic entity composition under zero-shot out-of-domain (OOD) generalization, where such dynamic changes cannot be anticipated beforehand. Our empirical studies reveal that existing MARL methods suffer significant performance degradation and increased uncertainty in these scenarios. In response, we propose FlickerFusion, a novel OOD generalization method that acts as a universally applicable augmentation technique for MARL backbone methods. Our results show that FlickerFusion not only achieves superior inference rewards but also uniquely reduces uncertainty vis-\`a-vis the backbone, compared to existing methods. For standardized evaluation, we introduce MPEv2, an enhanced version of Multi Particle Environments (MPE), consisting of 12 benchmarks. Benchmarks, implementations, and trained models are organized and open-sourced at flickerfusion305.github.io, accompanied by ample demo video renderings.
Authors: Chen Hu, Hanchi Ren, Jingjing Deng, Xianghua Xie
Abstract: Unmanned Aerial Vehicle (UAV) swarms are increasingly deployed in dynamic, data-rich environments for applications such as environmental monitoring and surveillance. These scenarios demand efficient data processing while maintaining privacy and security, making Federated Learning (FL) a promising solution. FL allows UAVs to collaboratively train global models without sharing raw data, but challenges arise due to the non-Independent and Identically Distributed (non-IID) nature of the data collected by UAVs. In this study, we show an integration of the state-of-the-art FL methods to UAV Swarm application and invetigate the performance of multiple aggregation methods (namely FedAvg, FedProx, FedOpt, and MOON) with a particular focus on tackling non-IID on a variety of datasets, specifically MNIST for baseline performance, CIFAR10 for natural object classification, EuroSAT for environment monitoring, and CelebA for surveillance. These algorithms were selected to cover improved techniques on both client-side updates and global aggregation. Results show that while all algorithms perform comparably on IID data, their performance deteriorates significantly under non-IID conditions. FedProx demonstrated the most stable overall performance, emphasising the importance of regularising local updates in non-IID environments to mitigate drastic deviations in local models.
Authors: Kirill Lukyanov, Andrew Perminov, Denis Turdakov, Mikhail Pautov
Abstract: The vulnerability of artificial neural networks to adversarial perturbations in the black-box setting is widely studied in the literature. The majority of attack methods to construct these perturbations suffer from an impractically large number of queries required to find an adversarial example. In this work, we focus on knowledge distillation as an approach to conduct transfer-based black-box adversarial attacks and propose an iterative training of the surrogate model on an expanding dataset. This work is the first, to our knowledge, to provide provable guarantees on the success of knowledge distillation-based attack on classification neural networks: we prove that if the student model has enough learning capabilities, the attack on the teacher model is guaranteed to be found within the finite number of distillation iterations.
Authors: Hanlin Yang, Jian Yao, Weiming Liu, Qing Wang, Hanmin Qin, Hansheng Kong, Kirk Tang, Jiechao Xiong, Chao Yu, Kai Li, Junliang Xing, Hongwu Chen, Juchao Zhuo, Qiang Fu, Yang Wei, Haobo Fu
Abstract: Recovering a spectrum of diverse policies from a set of expert trajectories is an important research topic in imitation learning. After determining a latent style for a trajectory, previous diverse policies recovering methods usually employ a vanilla behavioral cloning learning objective conditioned on the latent style, treating each state-action pair in the trajectory with equal importance. Based on an observation that in many scenarios, behavioral styles are often highly relevant with only a subset of state-action pairs, this paper presents a new principled method in diverse polices recovery. In particular, after inferring or assigning a latent style for a trajectory, we enhance the vanilla behavioral cloning by incorporating a weighting mechanism based on pointwise mutual information. This additional weighting reflects the significance of each state-action pair's contribution to learning the style, thus allowing our method to focus on state-action pairs most representative of that style. We provide theoretical justifications for our new objective, and extensive empirical evaluations confirm the effectiveness of our method in recovering diverse policies from expert data.
Authors: Kejia Fan, Jiaxu Li, Songning Lai, Linpu Lv, Anfeng Liu, Jianheng Tang, Houbing Herbert Song, Huiping Zhuang
Abstract: Class-incremental Learning (CIL) in Time Series Classification (TSC) aims to incrementally train models using the streaming time series data that arrives continuously. The main problem in this scenario is catastrophic forgetting, i.e., training models with new samples inevitably leads to the forgetting of previously learned knowledge. Among existing methods, the replay-based methods achieve satisfactory performance but compromise privacy, while exemplar-free methods protect privacy but suffer from low accuracy. However, more critically, owing to their reliance on gradient-based update techniques, these existing methods fundamentally cannot solve the catastrophic forgetting problem. In TSC scenarios with continuously arriving data and temporally shifting distributions, these methods become even less practical. In this paper, we propose a Time Series Analytic Continual Learning framework, called TS-ACL. Inspired by analytical learning, TS-ACL transforms neural network updates into gradient-free linear regression problems, thereby fundamentally mitigating catastrophic forgetting. Specifically, employing a pre-trained and frozen feature extraction encoder, TS-ACL only needs to update its analytic classifier recursively in a lightweight manner that is highly suitable for real-time applications and large-scale data processing. Additionally, we theoretically demonstrate that the model obtained recursively through the TS-ACL is exactly equivalent to a model trained on the complete dataset in a centralized manner, thereby establishing the property of absolute knowledge memory. Extensive experiments validate the superior performance of our TS-ACL.
Authors: Shreya Arvind, Rishabh Pomaje, Rajshekhar V Bhat
Abstract: This paper presents a novel approach to solving convex optimization problems by leveraging the fact that, under certain regularity conditions, any set of primal or dual variables satisfying the Karush-Kuhn-Tucker (KKT) conditions is necessary and sufficient for optimality. Similar to Theory-Trained Neural Networks (TTNNs), the parameters of the convex optimization problem are input to the neural network, and the expected outputs are the optimal primal and dual variables. A choice for the loss function in this case is a loss, which we refer to as the KKT Loss, that measures how well the network's outputs satisfy the KKT conditions. We demonstrate the effectiveness of this approach using a linear program as an example. For this problem, we observe that minimizing the KKT Loss alone outperforms training the network with a weighted sum of the KKT Loss and a Data Loss (the mean-squared error between the ground truth optimal solutions and the network's output). Moreover, minimizing only the Data Loss yields inferior results compared to those obtained by minimizing the KKT Loss. While the approach is promising, the obtained primal and dual solutions are not sufficiently close to the ground truth optimal solutions. In the future, we aim to develop improved models to obtain solutions closer to the ground truth and extend the approach to other problem classes.
Authors: Shiyan Hu, Kai Zhao, Xiangfei Qiu, Yang Shu, Jilin Hu, Bin Yang, Chenjuan Guo
Abstract: Many methods have been proposed for unsupervised time series anomaly detection. Despite some progress, research on predicting future anomalies is still relatively scarce. Predicting anomalies is particularly challenging due to the diverse reaction time and the lack of labeled data. To address these challenges, we propose MultiRC to integrate reconstructive and contrastive learning for joint learning of anomaly prediction and detection, with multi-scale structure and adaptive dominant period mask to deal with the diverse reaction time. MultiRC also generates negative samples to provide essential training momentum for the anomaly prediction tasks and prevent model degradation. We evaluate seven benchmark datasets from different fields. For both anomaly prediction and detection tasks, MultiRC outperforms existing state-of-the-art methods.
Authors: Raghav Bongole, Amaury Gouverneur, Borja Rodr\'iguez-G\'alvez, Tobias J. Oechtering, Mikael Skoglund
Abstract: We study agents acting in an unknown environment where the agent's goal is to find a robust policy. We consider robust policies as policies that achieve high cumulative rewards for all possible environments. To this end, we consider agents minimizing the maximum regret over different environment parameters, leading to the study of minimax regret. This research focuses on deriving information-theoretic bounds for minimax regret in Markov Decision Processes (MDPs) with a finite time horizon. Building on concepts from supervised learning, such as minimum excess risk (MER) and minimax excess risk, we use recent bounds on the Bayesian regret to derive minimax regret bounds. Specifically, we establish minimax theorems and use bounds on the Bayesian regret to perform minimax regret analysis using these minimax theorems. Our contributions include defining a suitable minimax regret in the context of MDPs, finding information-theoretic bounds for it, and applying these bounds in various scenarios.
Authors: Arijit Das
Abstract: Training LLMs presents significant memory challenges due to growing size of data, weights, and optimizer states. Techniques such as data and model parallelism, gradient checkpointing, and offloading strategies address this issue but are often infeasible due to hardware constraints. To mitigate memory usage, alternative methods like Parameter-Efficient-Fine-Tuning (PEFT) and GaLore approximate weights or optimizer states. PEFT methods, such as LoRA, have gained popularity for fine-tuning LLMs, though they require a full-rank warm start. In contrast, GaLore allows full-parameter learning while being more memory-efficient. This work introduces Natural GaLore, a simple drop in replacement for AdamW, which efficiently applies the inverse Empirical Fisher Information Matrix to low-rank gradients using Woodbury's Identity. We demonstrate that incorporating second-order information speeds up optimization significantly, especially when the iteration budget is limited. Empirical pretraining on 60M, 130M, 350M, and 1.1B parameter Llama models on C4 data demonstrate significantly lower perplexity over GaLore without additional memory overhead. By fine-tuning RoBERTa on the GLUE benchmark using Natural GaLore, we demonstrate significant reduction in gap 86.05% vs 86.28% for full-finetuning. Furthermore, fine-tuning the TinyLlama 1.1B model for function calling using the TinyAgent framework shows that Natural GaLore achieving 83.09% accuracy on the TinyAgent dataset, significantly outperforms 16-bit LoRA at 80.06% and even surpasses GPT4-Turbo by 4%, all while using 30% less memory. All code to reproduce the results are available at: https://github.com/selfsupervised-ai/Natural-GaLore.git
URLs: https://github.com/selfsupervised-ai/Natural-GaLore.git
Authors: Shiyu Wang, Jiawei Li, Xiaoming Shi, Zhou Ye, Baichuan Mo, Wenze Lin, Shengtong Ju, Zhixuan Chu, Ming Jin
Abstract: Time series analysis plays a critical role in numerous applications, supporting tasks such as forecasting, classification, anomaly detection, and imputation. In this work, we present the time series pattern machine (TSPM), a model designed to excel in a broad range of time series tasks through powerful representation and pattern extraction capabilities. Traditional time series models often struggle to capture universal patterns, limiting their effectiveness across diverse tasks. To address this, we define multiple scales in the time domain and various resolutions in the frequency domain, employing various mixing strategies to extract intricate, task-adaptive time series patterns. Specifically, we introduce a general-purpose TSPM that processes multi-scale time series using (1) multi-resolution time imaging (MRTI), (2) time image decomposition (TID), (3) multi-scale mixing (MCM), and (4) multi-resolution mixing (MRM) to extract comprehensive temporal patterns. MRTI transforms multi-scale time series into multi-resolution time images, capturing patterns across both temporal and frequency domains. TID leverages dual-axis attention to extract seasonal and trend patterns, while MCM hierarchically aggregates these patterns across scales. MRM adaptively integrates all representations across resolutions. This method achieves state-of-the-art performance across 8 time series analytical tasks, consistently surpassing both general-purpose and task-specific models. Our work marks a promising step toward the next generation of TSPMs, paving the way for further advancements in time series analysis.
Authors: Shogo Iwazaki, Shion Takeno
Abstract: This paper studies a non-stationary kernelized bandit (KB) problem, also called time-varying Bayesian optimization, where one seeks to minimize the regret under an unknown reward function that varies over time. In particular, we focus on a near-optimal algorithm whose regret upper bound matches the regret lower bound. For this goal, we show the first algorithm-independent regret lower bound for non-stationary KB with squared exponential and Mat\'ern kernels, which reveals that an existing optimization-based KB algorithm with slight modification is near-optimal. However, this existing algorithm suffers from feasibility issues due to its huge computational cost. Therefore, we propose a novel near-optimal algorithm called restarting phased elimination with random permutation (R-PERP), which bypasses the huge computational cost. A technical key point is the simple permutation procedures of query candidates, which enable us to derive a novel tighter confidence bound tailored to the non-stationary problems.
Authors: Zhenpeng Su, Xing Wu, Zijia Lin, Yizhe Xiong, Minxuan Lv, Guangyuan Ma, Hui Chen, Songlin Hu, Guiguang Ding
Abstract: Large language models (LLM) have been attracting much attention from the community recently, due to their remarkable performance in all kinds of downstream tasks. According to the well-known scaling law, scaling up a dense LLM enhances its capabilities, but also significantly increases the computational complexity. Mixture-of-Experts (MoE) models address that by allowing the model size to grow without substantially raising training or inference costs. Yet MoE models face challenges regarding knowledge sharing among experts, making their performance somehow sensitive to routing accuracy. To tackle that, previous works introduced shared experts and combined their outputs with those of the top $K$ routed experts in an ``addition'' manner. In this paper, inspired by collective matrix factorization to learn shared knowledge among data, we propose CartesianMoE, which implements more effective knowledge sharing among experts in more like a ``multiplication'' manner. Extensive experimental results indicate that CartesianMoE outperforms previous MoE models for building LLMs, in terms of both perplexity and downstream task performance. And we also find that CartesianMoE achieves better expert routing robustness.
Authors: Pavel Ryt\'i\v{r}, Ale\v{s} Wodecki, Georgios Korpas, Jakub Mare\v{c}ek
Abstract: Causal learning from data has received much attention in recent years. One way of capturing causal relationships is by utilizing Bayesian networks. There, one recovers a weighted directed acyclic graph, in which random variables are represented by vertices, and the weights associated with each edge represent the strengths of the causal relationships between them. This concept is extended to capture dynamic effects by introducing a dependency on past data, which may be captured by the structural equation model, which is utilized in the present contribution to formulate a score-based learning approach. A mixed-integer quadratic program is formulated and an algorithmic solution proposed, in which the pre-generation of exponentially many acyclicity constraints is avoided by utilizing the so-called branch-and-cut ("lazy constraint") method. Comparing the novel approach to the state of the art, we show that the proposed approach turns out to produce excellent results when applied to small and medium-sized synthetic instances of up to 25 time-series. Lastly, two interesting applications in bio-science and finance, to which the method is directly applied, further stress the opportunities in developing highly accurate, globally convergent solvers that can handle modest instances.
Authors: Thomas Robert, Mher Safaryan, Ionut-Vlad Modoranu, Dan Alistarh
Abstract: We introduce LDAdam, a memory-efficient optimizer for training large models, that performs adaptive optimization steps within lower dimensional subspaces, while consistently exploring the full parameter space during training. This strategy keeps the optimizer's memory footprint to a fraction of the model size. LDAdam relies on a new projection-aware update rule for the optimizer states that allows for transitioning between subspaces, i.e., estimation of the statistics of the projected gradients. To mitigate the errors due to low-rank projection, LDAdam integrates a new generalized error feedback mechanism, which explicitly accounts for both gradient and optimizer state compression. We prove the convergence of LDAdam under standard assumptions, and show that LDAdam allows for accurate and efficient fine-tuning and pre-training of language models.
Authors: Ronglong Fang, Yuesheng Xu
Abstract: Deep neural networks (DNNs) suffer from the spectral bias, wherein DNNs typically exhibit a tendency to prioritize the learning of lower-frequency components of a function, struggling to capture its high-frequency features. This paper is to address this issue. Notice that a function having only low frequency components may be well-represented by a shallow neural network (SNN), a network having only a few layers. By observing that composition of low frequency functions can effectively approximate a high-frequency function, we propose to learn a function containing high-frequency components by composing several SNNs, each of which learns certain low-frequency information from the given data. We implement the proposed idea by exploiting the multi-grade deep learning (MGDL) model, a recently introduced model that trains a DNN incrementally, grade by grade, a current grade learning from the residue of the previous grade only an SNN composed with the SNNs trained in the preceding grades as features. We apply MGDL to synthetic, manifold, colored images, and MNIST datasets, all characterized by presence of high-frequency features. Our study reveals that MGDL excels at representing functions containing high-frequency information. Specifically, the neural networks learned in each grade adeptly capture some low-frequency information, allowing their compositions with SNNs learned in the previous grades effectively representing the high-frequency features. Our experimental results underscore the efficacy of MGDL in addressing the spectral bias inherent in DNNs. By leveraging MGDL, we offer insights into overcoming spectral bias limitation of DNNs, thereby enhancing the performance and applicability of deep learning models in tasks requiring the representation of high-frequency information. This study confirms that the proposed method offers a promising solution to address the spectral bias of DNNs.
Authors: Swagatam Haldar, Christoph Stein-Thoeringer, Vadim Borisov
Abstract: Understanding the complex interactions within the microbiome is crucial for developing effective diagnostic and therapeutic strategies. Traditional machine learning models often lack interpretability, which is essential for clinical and biological insights. This paper explores the application of symbolic regression (SR) to microbiome relative abundance data, with a focus on colorectal cancer (CRC). SR, known for its high interpretability, is compared against traditional machine learning models, e.g., random forest, gradient boosting decision trees. These models are evaluated based on performance metrics such as F1 score and accuracy. We utilize 71 studies encompassing, from various cohorts, over 10,000 samples across 749 species features. Our results indicate that SR not only competes reasonably well in terms of predictive performance, but also excels in model interpretability. SR provides explicit mathematical expressions that offer insights into the biological relationships within the microbiome, a crucial advantage for clinical and biological interpretation. Our experiments also show that SR can help understand complex models like XGBoost via knowledge distillation. To aid in reproducibility and further research, we have made the code openly available at https://github.com/swag2198/microbiome-symbolic-regression .
URLs: https://github.com/swag2198/microbiome-symbolic-regression
Authors: Xinyi Zhou, Xing Li, Yingzhao Lian, Yiwen Wang, Lei Chen, Mingxuan Yuan, Jianye Hao, Guangyong Chen, Pheng Ann Heng
Abstract: We introduce SeaDAG, a semi-autoregressive diffusion model for conditional generation of Directed Acyclic Graphs (DAGs). Considering their inherent layer-wise structure, we simulate layer-wise autoregressive generation by designing different denoising speed for different layers. Unlike conventional autoregressive generation that lacks a global graph structure view, our method maintains a complete graph structure at each diffusion step, enabling operations such as property control that require the full graph structure. Leveraging this capability, we evaluate the DAG properties during training by employing a graph property decoder. We explicitly train the model to learn graph conditioning with a condition loss, which enhances the diffusion model's capacity to generate graphs that are both realistic and aligned with specified properties. We evaluate our method on two representative conditional DAG generation tasks: (1) circuit generation from truth tables, where precise DAG structures are crucial for realizing circuit functionality, and (2) molecule generation based on quantum properties. Our approach demonstrates promising results, generating high-quality and realistic DAGs that closely align with given conditions.
Authors: Lele Zheng, Yang Cao, Renhe Jiang, Kenjiro Taura, Yulong Shen, Sheng Li, Masatoshi Yoshikawa
Abstract: Recent works show that sensitive user data can be reconstructed from gradient updates, breaking the key privacy promise of federated learning. While success was demonstrated primarily on image data, these methods do not directly transfer to other domains, such as spatiotemporal data. To understand privacy risks in spatiotemporal federated learning, we first propose Spatiotemporal Gradient Inversion Attack (ST-GIA), a gradient attack algorithm tailored to spatiotemporal data that successfully reconstructs the original location from gradients. Furthermore, the absence of priors in attacks on spatiotemporal data has hindered the accurate reconstruction of real client data. To address this limitation, we propose ST-GIA+, which utilizes an auxiliary language model to guide the search for potential locations, thereby successfully reconstructing the original data from gradients. In addition, we design an adaptive defense strategy to mitigate gradient inversion attacks in spatiotemporal federated learning. By dynamically adjusting the perturbation levels, we can offer tailored protection for varying rounds of training data, thereby achieving a better trade-off between privacy and utility than current state-of-the-art methods. Through intensive experimental analysis on three real-world datasets, we reveal that the proposed defense strategy can well preserve the utility of spatiotemporal federated learning with effective security protection.
Authors: Polina Turishcheva, Laura Hansel, Martin Ritzert, Marissa A. Weis, Alexander S. Ecker
Abstract: Driven by advances in recording technology, large-scale high-dimensional datasets have emerged across many scientific disciplines. Especially in biology, clustering is often used to gain insights into the structure of such datasets, for instance to understand the organization of different cell types. However, clustering is known to scale poorly to high dimensions, even though the exact impact of dimensionality is unclear as current benchmark datasets are mostly two-dimensional. Here we propose MNIST-Nd, a set of synthetic datasets that share a key property of real-world datasets, namely that individual samples are noisy and clusters do not perfectly separate. MNIST-Nd is obtained by training mixture variational autoencoders with 2 to 64 latent dimensions on MNIST, resulting in six datasets with comparable structure but varying dimensionality. It thus offers the chance to disentangle the impact of dimensionality on clustering. Preliminary common clustering algorithm benchmarks on MNIST-Nd suggest that Leiden is the most robust for growing dimensions.
Authors: Kang Zhao, Tao Yuan, Han Bao, Zhenfeng Su, Chang Gao, Zhaofeng Sun, Zichen Liang, Liping Jing, Jianfei Chen
Abstract: To date, 2:4 sparsity has stood as the only sparse pattern that can be accelerated using sparse tensor cores on GPUs. In practice, 2:4 sparsity often possesses low actual speedups ($\leq 1.3$) and requires fixed sparse ratios, meaning that other ratios, such as 4:8, 8:16, or those exceeding 50% sparsity, do not incur any speedups on GPUs. Recent studies suggest that V:N:M sparsity is promising in addressing these limitations of 2:4 sparsity. However, regarding accuracy, the effects of V:N:M sparsity on broader Transformer models, such as vision Transformers and large language models (LLMs), are largely unexamined. Moreover, Some specific issues related to V:N:M sparsity, such as how to select appropriate V and M values, remain unresolved. In this study, we thoroughly investigate the application of V:N:M sparsity in vision models and LLMs across multiple tasks, from pertaining to downstream tasks. We propose three key approaches to enhance the applicability and accuracy of V:N:M-sparse Transformers, including heuristic V and M selection, V:N:M-specific channel permutation, and three-staged LoRA training techniques. Experimental results show that, with our methods, the DeiT-small achieves lossless accuracy at 64:2:5 sparsity, while the DeiT-base maintains accuracy even at 64:2:8 sparsity. In addition, the fine-tuned LLama2-7B at 64:2:5 sparsity performs comparably or better than training-free 2:4 sparse alternatives on downstream tasks. More importantly, V:N:M-sparse Transformers offer a wider range of speedup-accuracy trade-offs compared to 2:4 sparsity. Overall, our exploration largely facilitates the V:N:M sparsity to act as a truly effective acceleration solution for Transformers in cost-sensitive inference scenarios.
Authors: Fan Yang, Xingyue Huang
Abstract: Line graph transformation has been widely studied in graph theory, where each node in a line graph corresponds to an edge in the original graph. This has inspired a series of graph neural networks (GNNs) applied to transformed line graphs, which have proven effective in various graph representation learning tasks. However, there is limited theoretical study on how line graph transformation affects the expressivity of GNN models. In this study, we focus on two types of graphs known to be challenging to the Weisfeiler-Leman (WL) tests: Cai-F\"urer-Immerman (CFI) graphs and strongly regular graphs, and show that applying line graph transformation helps exclude these challenging graph properties, thus potentially assist WL tests in distinguishing these graphs. We empirically validate our findings by conducting a series of experiments that compare the accuracy and efficiency of graph isomorphism tests and GNNs on both line-transformed and original graphs across these graph structure types.
Authors: Xilin He, Jingyu Hu, Qinliang Lin, Cheng Luo, Weicheng Xie, Siyang Song, Muhammad Haris Khan, Linlin Shen
Abstract: Domain generalization methods aim to learn transferable knowledge from source domains that can generalize well to unseen target domains. Recent studies show that neural networks frequently suffer from a simplicity-biased learning behavior which leads to over-reliance on specific frequency sets, namely as frequency shortcuts, instead of semantic information, resulting in poor generalization performance. Despite previous data augmentation techniques successfully enhancing generalization performances, they intend to apply more frequency shortcuts, thereby causing hallucinations of generalization improvement. In this paper, we aim to prevent such learning behavior of applying frequency shortcuts from a data-driven perspective. Given the theoretical justification of models' biased learning behavior on different spatial frequency components, which is based on the dataset frequency properties, we argue that the learning behavior on various frequency components could be manipulated by changing the dataset statistical structure in the Fourier domain. Intuitively, as frequency shortcuts are hidden in the dominant and highly dependent frequencies of dataset structure, dynamically perturbating the over-reliance frequency components could prevent the application of frequency shortcuts. To this end, we propose two effective data augmentation modules designed to collaboratively and adaptively adjust the frequency characteristic of the dataset, aiming to dynamically influence the learning behavior of the model and ultimately serving as a strategy to mitigate shortcut learning. Code is available at AdvFrequency (https://github.com/C0notSilly/AdvFrequency).
Authors: Robin Th\'eriault, Francesco Tosello, Daniele Tantari
Abstract: Restricted Boltzmann machines (RBM) are generative models capable to learn data with a rich underlying structure. We study the teacher-student setting where a student RBM learns structured data generated by a teacher RBM. The amount of structure in the data is controlled by adjusting the number of hidden units of the teacher and the correlations in the rows of the weights, a.k.a. patterns. In the absence of correlations, we validate the conjecture that the performance is independent of the number of teacher patters and hidden units of the student RBMs, and we argue that the teacher-student setting can be used as a toy model for studying the lottery ticket hypothesis. Beyond this regime, we find that the critical amount of data required to learn the teacher patterns decreases with both their number and correlations. In both regimes, we find that, even with an relatively large dataset, it becomes impossible to learn the teacher patterns if the inference temperature used for regularization is kept too low. In our framework, the student can learn teacher patterns one-to-one or many-to-one, generalizing previous findings about the teacher-student setting with two hidden units to any arbitrary finite number of hidden units.
Authors: Mostafa Hussien, Mahmoud Afifi, Kim Khoa Nguyen, Mohamed Cheriet
Abstract: Recent advancements have scaled neural networks to unprecedented sizes, achieving remarkable performance across a wide range of tasks. However, deploying these large-scale models on resource-constrained devices poses significant challenges due to substantial storage and computational requirements. Neural network pruning has emerged as an effective technique to mitigate these limitations by reducing model size and complexity. In this paper, we introduce an intuitive and interpretable pruning method based on activation statistics, rooted in information theory and statistical analysis. Our approach leverages the statistical properties of neuron activations to identify and remove weights with minimal contributions to neuron outputs. Specifically, we build a distribution of weight contributions across the dataset and utilize its parameters to guide the pruning process. Furthermore, we propose a Pruning-aware Training strategy that incorporates an additional regularization term to enhance the effectiveness of our pruning method. Extensive experiments on multiple datasets and network architectures demonstrate that our method consistently outperforms several baseline and state-of-the-art pruning techniques.
Authors: Anthony Bazhenov, Pahan Dewasurendra, Giri P. Krishnan, Jean Erik Delanois
Abstract: Artificial neural networks (ANNs) show limited performance with scarce or imbalanced training data and face challenges with continuous learning, such as forgetting previously learned data after new tasks training. In contrast, the human brain can learn continuously and from just a few examples. This research explores the impact of 'sleep', an unsupervised phase incorporating stochastic activation with local Hebbian learning rules, on ANNs trained incrementally with limited and imbalanced datasets, specifically MNIST and Fashion MNIST. We discovered that introducing a sleep phase significantly enhanced accuracy in models trained with limited data. When a few tasks were trained sequentially, sleep replay not only rescued previously learned information that had been catastrophically forgetting following new task training but often enhanced performance in prior tasks, especially those trained with limited data. This study highlights the multifaceted role of sleep replay in augmenting learning efficiency and facilitating continual learning in ANNs.
Authors: Suman Sapkota
Abstract: Artificial Neural Networks of varying architectures are generally paired with affine transformation at the core. However, we find dot product neurons with global influence less interpretable as compared to local influence of euclidean distance (as used in Radial Basis Function Network). In this work, we explore the generalization of dot product neurons to $l^p$-norm, metrics, and beyond. We find that metrics as transform performs similarly to affine transform when used in MultiLayer Perceptron or Convolutional Neural Network. Moreover, we explore various properties of Metrics, compare it with Affine, and present multiple cases where metrics seem to provide better interpretability. We develop an interpretable local dictionary based Neural Networks and use it to understand and reject adversarial examples.
Authors: Liam Pavlovic, David M. Rosen
Abstract: Stein variational inference (SVI) is a sample-based approximate Bayesian inference technique that generates a sample set by jointly optimizing the samples' locations to minimize an information-theoretic measure of discrepancy with the target probability distribution. SVI thus provides a fast and significantly more sample-efficient approach to Bayesian inference than traditional (random-sampling-based) alternatives. However, the optimization techniques employed in existing SVI methods struggle to address problems in which the target distribution is high-dimensional, poorly-conditioned, or non-convex, which severely limits the range of their practical applicability. In this paper, we propose a novel trust-region optimization approach for SVI that successfully addresses each of these challenges. Our method builds upon prior work in SVI by leveraging conditional independences in the target distribution (to achieve high-dimensional scaling) and second-order information (to address poor conditioning), while additionally providing an effective adaptive step control procedure, which is essential for ensuring convergence on challenging non-convex optimization problems. Experimental results show our method achieves superior numerical performance, both in convergence rate and sample accuracy, and scales better in high-dimensional distributions, than previous SVI techniques.
Authors: Yuchen Cao, Jianglai Dai, Zhongyan Wang, Yeyubei Zhang, Xiaorui Shen, Yunchong Liu, Yexin Tian
Abstract: The global rise in depression necessitates innovative detection methods for early intervention. Social media provides a unique opportunity to identify depression through user-generated posts. This systematic review evaluates machine learning (ML) models for depression detection on social media, focusing on biases and methodological challenges throughout the ML lifecycle. A search of PubMed, IEEE Xplore, and Google Scholar identified 47 relevant studies published after 2010. The Prediction model Risk Of Bias ASsessment Tool (PROBAST) was utilized to assess methodological quality and risk of bias. Significant biases impacting model reliability and generalizability were found. There is a predominant reliance on Twitter (63.8%) and English-language content (over 90%), with most studies focusing on users from the United States and Europe. Non-probability sampling methods (approximately 80%) limit representativeness. Only 23% of studies explicitly addressed linguistic nuances like negations, crucial for accurate sentiment analysis. Inconsistent hyperparameter tuning was observed, with only 27.7% properly tuning models. About 17% did not adequately partition data into training, validation, and test sets, risking overfitting. While 74.5% used appropriate evaluation metrics for imbalanced data, others relied on accuracy without addressing class imbalance, potentially skewing results. Reporting transparency varied, often lacking critical methodological details. These findings highlight the need to diversify data sources, standardize preprocessing protocols, ensure consistent model development practices, address class imbalance, and enhance reporting transparency. By overcoming these challenges, future research can develop more robust and generalizable ML models for depression detection on social media, contributing to improved mental health outcomes globally.
Authors: Junjie Oscar Yin, Alexander M. Rush
Abstract: Data selection can reduce the amount of training data needed to finetune LLMs; however, the efficacy of data selection scales directly with its compute. Motivated by the practical challenge of compute-constrained finetuning, we consider the setting in which both the cost of selecting data and training are budgeted for. We first formalize the problem of data selection with a cost-aware utility function, and model the data selection problem as trading off initial-selection cost for training gain. We run a comprehensive sweep of experiments across multiple tasks, varying compute budget by scaling finetuning tokens, model sizes, and data selection compute. These experiments show the validity of this model in real-world experiments. Interestingly we find that many powerful data selection methods are almost never compute-optimal, and that cheaper data selection alternatives dominate both from a theoretical and empirical perspective.
Authors: Valentyn Boreiko, Alexander Panfilov, Vaclav Voracek, Matthias Hein, Jonas Geiping
Abstract: A plethora of jailbreaking attacks have been proposed to obtain harmful responses from safety-tuned LLMs. In their original settings, these methods all largely succeed in coercing the target output, but their attacks vary substantially in fluency and computational effort. In this work, we propose a unified threat model for the principled comparison of these methods. Our threat model combines constraints in perplexity, measuring how far a jailbreak deviates from natural text, and computational budget, in total FLOPs. For the former, we build an N-gram model on 1T tokens, which, in contrast to model-based perplexity, allows for an LLM-agnostic and inherently interpretable evaluation. We adapt popular attacks to this new, realistic threat model, with which we, for the first time, benchmark these attacks on equal footing. After a rigorous comparison, we not only find attack success rates against safety-tuned modern models to be lower than previously presented but also find that attacks based on discrete optimization significantly outperform recent LLM-based attacks. Being inherently interpretable, our threat model allows for a comprehensive analysis and comparison of jailbreak attacks. We find that effective attacks exploit and abuse infrequent N-grams, either selecting N-grams absent from real-world text or rare ones, e.g. specific to code datasets.
Authors: Santhosh Karnik, Anna Veselovska, Mark Iwen, Felix Krahmer
Abstract: We provide a rigorous analysis of implicit regularization in an overparametrized tensor factorization problem beyond the lazy training regime. For matrix factorization problems, this phenomenon has been studied in a number of works. A particular challenge has been to design universal initialization strategies which provably lead to implicit regularization in gradient-descent methods. At the same time, it has been argued by Cohen et. al. 2016 that more general classes of neural networks can be captured by considering tensor factorizations. However, in the tensor case, implicit regularization has only been rigorously established for gradient flow or in the lazy training regime. In this paper, we prove the first tensor result of its kind for gradient descent rather than gradient flow. We focus on the tubal tensor product and the associated notion of low tubal rank, encouraged by the relevance of this model for image data. We establish that gradient descent in an overparametrized tensor factorization model with a small random initialization exhibits an implicit bias towards solutions of low tubal rank. Our theoretical findings are illustrated in an extensive set of numerical simulations show-casing the dynamics predicted by our theory as well as the crucial role of using a small random initialization.
Authors: Nick Rittler, Kamalika Chaudhuri
Abstract: Generative models at times produce "invalid" outputs, such as images with generation artifacts and unnatural sounds. Validity-constrained distribution learning attempts to address this problem by requiring that the learned distribution have a provably small fraction of its mass in invalid parts of space -- something which standard loss minimization does not always ensure. To this end, a learner in this model can guide the learning via "validity queries", which allow it to ascertain the validity of individual examples. Prior work on this problem takes a worst-case stance, showing that proper learning requires an exponential number of validity queries, and demonstrating an improper algorithm which -- while generating guarantees in a wide-range of settings -- makes an atypical polynomial number of validity queries. In this work, we take a first step towards characterizing regimes where guaranteeing validity is easier than in the worst-case. We show that when the data distribution lies in the model class and the log-loss is minimized, the number of samples required to ensure validity has a weak dependence on the validity requirement. Additionally, we show that when the validity region belongs to a VC-class, a limited number of validity queries are often sufficient.
Authors: Yi Ding, Bolian Li, Ruqi Zhang
Abstract: Vision Language Models (VLMs) have become essential backbones for multimodal intelligence, yet significant safety challenges limit their real-world application. While textual inputs are often effectively safeguarded, adversarial visual inputs can easily bypass VLM defense mechanisms. Existing defense methods are either resource-intensive, requiring substantial data and compute, or fail to simultaneously ensure safety and usefulness in responses. To address these limitations, we propose a novel two-phase inference-time alignment framework, Evaluating Then Aligning (ETA): 1) Evaluating input visual contents and output responses to establish a robust safety awareness in multimodal settings, and 2) Aligning unsafe behaviors at both shallow and deep levels by conditioning the VLMs' generative distribution with an interference prefix and performing sentence-level best-of-N to search the most harmless and helpful generation paths. Extensive experiments show that ETA outperforms baseline methods in terms of harmlessness, helpfulness, and efficiency, reducing the unsafe rate by 87.5% in cross-modality attacks and achieving 96.6% win-ties in GPT-4 helpfulness evaluation. The code is publicly available at https://github.com/DripNowhy/ETA.
Authors: Jiahao Wang, Amer Shalaby
Abstract: Public transit systems play a crucial role in providing efficient and sustainable transportation options in urban areas. However, these systems face various challenges in meeting commuters' needs. On the other hand, despite the rapid development of Large Language Models (LLMs) worldwide, their integration into transit systems remains relatively unexplored. The objective of this paper is to explore the utilization of LLMs in the public transit system, with a specific focus on improving the customers' experience and transit staff performance. We present a general framework for developing LLM applications in transit systems, wherein the LLM serves as the intermediary for information communication between natural language content and the resources within the database. In this context, the LLM serves a multifaceted role, including understanding users' requirements, retrieving data from the dataset in response to user queries, and tailoring the information to align with the users' specific needs. Three transit LLM applications are presented: Tweet Writer, Trip Advisor, and Policy Navigator. Tweet Writer automates updates to the transit system alerts on social media, Trip Advisor offers customized transit trip suggestions, and Policy Navigator provides clear and personalized answers to policy queries. Leveraging LLMs in these applications enhances seamless communication with their capabilities of understanding and generating human-like languages. With the help of these three LLM transit applications, transit system media personnel can provide system updates more efficiently, and customers can access travel information and policy answers in a more user-friendly manner.
Authors: Lucas Heublein, Tobias Feigl, Alexander R\"ugamer, Felix Ott
Abstract: The accuracy of global navigation satellite system (GNSS) receivers is significantly compromised by interference from jamming devices. Consequently, the detection of these jammers are crucial to mitigating such interference signals. However, robust classification of interference using machine learning (ML) models is challenging due to the lack of labeled data in real-world environments. In this paper, we propose an ML approach that achieves high generalization in classifying interference through orchestrated monitoring stations deployed along highways. We present a semi-supervised approach coupled with an uncertainty-based voting mechanism by combining Monte Carlo and Deep Ensembles that effectively minimizes the requirement for labeled training samples to less than 5% of the dataset while improving adaptability across varying environments. Our method demonstrates strong performance when adapted from indoor environments to real-world scenarios.
Authors: Zhengzheng Tang
Abstract: This study introduces BrainTransformers, an innovative Large Language Model (LLM) implemented using Spiking Neural Networks (SNN). Our key contributions include: (1) designing SNN-compatible Transformer components such as SNNMatmul, SNNSoftmax, and SNNSiLU; (2) implementing an SNN approximation of the SiLU activation function; and (3) developing a Synapsis module to simulate synaptic plasticity. Our 3-billion parameter model, BrainTransformers-3B-Chat, demonstrates competitive performance across various benchmarks, including MMLU (63.2), BBH (54.1), ARC-C (54.3), and GSM8K (76.3), while potentially offering improved energy efficiency and biological plausibility. The model employs a three-stage training approach, including SNN-specific neuronal synaptic plasticity training. This research opens new avenues for brain-like AI systems in natural language processing and neuromorphic computing. Future work will focus on hardware optimization, developing specialized SNN fine-tuning tools, and exploring practical applications in energy-efficient computing environments.
Authors: Taewon Kim, Hyunjin Seo, Sungsoo Ahn, Eunho Yang
Abstract: Predicting the ground-state 3D molecular conformations from 2D molecular graphs is critical in computational chemistry due to its profound impact on molecular properties. Deep learning (DL) approaches have recently emerged as promising alternatives to computationally-heavy classical methods such as density functional theory (DFT). However, we discover that existing DL methods inadequately model inter-atomic forces, particularly for non-bonded atomic pairs, due to their naive usage of bonds and pairwise distances. Consequently, significant prediction errors occur for atoms with low degree (i.e., low coordination numbers) whose conformations are primarily influenced by non-bonded interactions. To address this, we propose REBIND, a novel framework that rewires molecular graphs by adding edges based on the Lennard-Jones potential to capture non-bonded interactions for low-degree atoms. Experimental results demonstrate that REBIND significantly outperforms state-of-the-art methods across various molecular sizes, achieving up to a 20\% reduction in prediction error.
Authors: Maciej Adamiak, Yulia Grinblat, Julian Psotta, Nir Fulman, Himshikhar Mazumdar, Shiyu Tang, Alexander Zipf
Abstract: This paper presents a method for detecting and estimating vehicle speeds using PlanetScope SuperDove satellite imagery, offering a scalable solution for global vehicle traffic monitoring. Conventional methods such as stationary sensors and mobile systems like UAVs are limited in coverage and constrained by high costs and legal restrictions. Satellite-based approaches provide broad spatial coverage but face challenges, including high costs, low frame rates, and difficulty detecting small vehicles in high-resolution imagery. We propose a Keypoint R-CNN model to track vehicle trajectories across RGB bands, leveraging band timing differences to estimate speed. Validation is performed using drone footage and GPS data covering highways in Germany and Poland. Our model achieved a Mean Average Precision of 0.53 and velocity estimation errors of approximately 3.4 m/s compared to GPS data. Results from drone comparison reveal underestimations, with average speeds of 112.85 km/h for satellite data versus 131.83 km/h from drone footage. While challenges remain with high-speed accuracy, this approach demonstrates the potential for scalable, daily traffic monitoring across vast areas, providing valuable insights into global traffic dynamics.
Authors: Paulo Luza Alves, Andr\'e Hochuli, Luiz Eduardo de Oliveira, Paulo Lisboa de Almeida
Abstract: When deploying large-scale machine learning models for smart city applications, such as image-based parking lot monitoring, data often must be sent to a central server to perform classification tasks. This is challenging for the city's infrastructure, where image-based applications require transmitting large volumes of data, necessitating complex network and hardware infrastructures to process the data. To address this issue in image-based parking space classification, we propose creating a robust ensemble of classifiers to serve as Teacher models. These Teacher models are distilled into lightweight and specialized Student models that can be deployed directly on edge devices. The knowledge is distilled to the Student models through pseudo-labeled samples generated by the Teacher model, which are utilized to fine-tune the Student models on the target scenario. Our results show that the Student models, with 26 times fewer parameters than the Teacher models, achieved an average accuracy of 96.6% on the target test datasets, surpassing the Teacher models, which attained an average accuracy of 95.3%.
Authors: Xiyu Zhai, Runlong Zhou, Liao Zhang, Simon Shaolei Du
Abstract: Transformer-based large language models (LLMs) have demonstrated surprisingly robust performance across a wide range of language-related tasks, including programming language understanding and generation. In this paper, we take the first steps towards a formal investigation of using transformers as compilers from an expressive power perspective. To this end, we introduce a representative programming language, Mini-Husky, which encapsulates key features of modern C-like languages. We show that if the input code sequence has a bounded depth in both the Abstract Syntax Tree (AST) and type inference (reasonable assumptions based on the clean code principle), then the number of parameters required by transformers depends only on the logarithm of the input sequence length to handle compilation tasks, such as AST construction, symbol resolution, and type analysis. A significant technical challenge stems from the fact that transformers operate at a low level, where each layer processes the input sequence as raw vectors without explicitly associating them with predefined structure or meaning. In contrast, high-level compiler tasks necessitate managing intricate relationships and structured program information. Our primary technical contribution is the development of a domain-specific language, Cybertron, which generates formal proofs of the transformer's expressive power, scaling to address compiler tasks. We further establish that recurrent neural networks (RNNs) require at least a linear number of parameters relative to the input sequence, leading to an exponential separation between transformers and RNNs. Finally, we empirically validate our theoretical results by comparing transformers and RNNs on compiler tasks within Mini-Husky.
Authors: Naoki Murata, Chieh-Hsin Lai, Yuhta Takida, Toshimitsu Uesaka, Bac Nguyen, Stefano Ermon, Yuki Mitsufuji
Abstract: Recent literature has effectively utilized diffusion models trained on continuous variables as priors for solving inverse problems. Notably, discrete diffusion models with discrete latent codes have shown strong performance, particularly in modalities suited for discrete compressed representations, such as image and motion generation. However, their discrete and non-differentiable nature has limited their application to inverse problems formulated in continuous spaces. This paper presents a novel method for addressing linear inverse problems by leveraging image-generation models based on discrete diffusion as priors. We overcome these limitations by approximating the true posterior distribution with a variational distribution constructed from categorical distributions and continuous relaxation techniques. Furthermore, we employ a star-shaped noise process to mitigate the drawbacks of traditional discrete diffusion models with absorbing states, demonstrating that our method performs comparably to continuous diffusion techniques. To the best of our knowledge, this is the first approach to use discrete diffusion model-based priors for solving image inverse problems.
Authors: Xiaofeng Tan
Abstract: For over half a century, computer-aided structural elucidation systems (CASE) for organic compounds have relied on complex expert systems with explicitly programmed algorithms. These systems are often computationally inefficient for complex compounds due to the vast chemical structural space that must be explored and filtered. In this study, we present a transformer based generative chemical language artificial intelligence (AI) model, an innovative end-to-end architecture designed to replace the logic and workflow of the classic CASE framework for ultra-fast and accurate spectroscopic-based structural elucidation. Our model employs an encoder-decoder architecture and self-attention mechanisms, similar to those in large language models, to directly generate the most probable chemical structures that match the input spectroscopic data. This approach demonstrates the potential of transformer based generative AI to accelerate traditional scientific problem-solving processes. The model's ability to iterate quickly based on new data highlights its potential for rapid advancements in structural elucidation.
Authors: A Mani
Abstract: The concepts of precision, and accuracy are domain and problem dependent. The simplified numeric hard and soft measures used in the fields of statistical learning, many types of machine learning, and binary or multiclass classification problems are known to be of limited use for understanding the meaningfulness of models or their relevance. Arguably, they are neither of patterns nor proofs. Further, there are no good measures or representations for analogous concepts in the cognition domain. In this research, the key issues are reflected upon, and a compositional knowledge representation approach in a minimalist general rough framework is proposed for the problem contexts. The latter is general enough to cover most application contexts, and may be applicable in the light of improved computational tools available.
Authors: Xinfu Li, Junying Zhang, Xindi Ma
Abstract: Backdoor defenses have recently become important in resisting backdoor attacks in deep neural networks (DNNs), where attackers implant backdoors into the DNN model by injecting backdoor samples into the training dataset. Although there are many defense methods to achieve backdoor detection for DNN inputs and backdoor elimination for DNN models, they still have not presented a clear explanation of the relationship between these two missions. In this paper, we use the features from the middle layer of the DNN model to analyze the difference between backdoor and benign samples and propose Backdoor Consistency, which indicates that at least one backdoor exists in the DNN model if the backdoor trigger is detected exactly on input. By analyzing the middle features, we design an effective and comprehensive backdoor defense method named BeniFul, which consists of two parts: a gray-box backdoor input detection and a white-box backdoor elimination. Specifically, we use the reconstruction distance from the Variational Auto-Encoder and model inference results to implement backdoor input detection and a feature distance loss to achieve backdoor elimination. Experimental results on CIFAR-10 and Tiny ImageNet against five state-of-the-art attacks demonstrate that our BeniFul exhibits a great defense capability in backdoor input detection and backdoor elimination.
Authors: Zixin Wang, Dong Gong, Sen Wang, Zi Huang, Yadan Luo
Abstract: In this work, we introduce Token Condensation as Adaptation (TCA), a training-free approach designed to mitigate distribution shifts encountered by vision-language models (VLMs) during test-time inference. TCA bridges distribution gaps at the patch level by condensing image tokens that exhibit low attentiveness to the
Authors: Yasin Hasanpoor, Bahram Tarvirdizadeh, Khalil Alipour, Mohammad Ghamari
Abstract: Our research introduces a groundbreaking approach to stress classification through Photoplethysmogram (PPG) signals. By combining Continuous Wavelet Transformation (CWT) with the proven VGG16 classifier, our method enhances stress assessment accuracy and reliability. Previous studies highlighted the importance of physiological signal analysis, yet precise stress classification remains a challenge. Our approach addresses this by incorporating robust data preprocessing with a Kalman filter and a sophisticated neural network architecture. Experimental results showcase exceptional performance, achieving a maximum training accuracy of 98% and maintaining an impressive average training accuracy of 96% across diverse stress scenarios. These results demonstrate the practicality and promise of our method in advancing stress monitoring systems and stress alarm sensors, contributing significantly to stress classification.
Authors: Emanuele Natale (CNRS, COATI, I3S, UniCA), Davide Ferr\'e (UniCA, CNRS, Inria, I3S), Giordano Giambartolomei (I3S, COMUE UCA, COATI), Fr\'ed\'eric Giroire (I3S, COMUE UCA, COATI), Frederik Mallmann-Trenn
Abstract: Considerable research efforts have recently been made to show that a random neural network $N$ contains subnetworks capable of accurately approximating any given neural network that is sufficiently smaller than $N$, without any training. This line of research, known as the Strong Lottery Ticket Hypothesis (SLTH), was originally motivated by the weaker Lottery Ticket Hypothesis, which states that a sufficiently large random neural network $N$ contains \emph{sparse} subnetworks that can be trained efficiently to achieve performance comparable to that of training the entire network $N$. Despite its original motivation, results on the SLTH have so far not provided any guarantee on the size of subnetworks. Such limitation is due to the nature of the main technical tool leveraged by these results, the Random Subset Sum (RSS) Problem. Informally, the RSS Problem asks how large a random i.i.d. sample $\Omega$ should be so that we are able to approximate any number in $[-1,1]$, up to an error of $ \epsilon$, as the sum of a suitable subset of $\Omega$. We provide the first proof of the SLTH in classical settings, such as dense and equivariant networks, with guarantees on the sparsity of the subnetworks. Central to our results, is the proof of an essentially tight bound on the Random Fixed-Size Subset Sum Problem (RFSS), a variant of the RSS Problem in which we only ask for subsets of a given size, which is of independent interest.
Authors: Mrinal Rawat, Hithesh Sankararaman, Victor Barres
Abstract: Deriving value from a conversational AI system depends on the capacity of a user to translate the prior knowledge into a configuration. In most cases, discovering the set of relevant turn-level speaker intents is often one of the key steps. Purely unsupervised algorithms provide a natural way to tackle discovery problems but make it difficult to incorporate constraints and only offer very limited control over the outcomes. Previous work has shown that semi-supervised (deep) clustering techniques can allow the system to incorporate prior knowledge and constraints in the intent discovery process. However they did not address how to allow for control through human feedback. In our Controllable Discovery of Intents (CDI) framework domain and prior knowledge are incorporated using a sequence of unsupervised contrastive learning on unlabeled data followed by fine-tuning on partially labeled data, and finally iterative refinement of clustering and representations through repeated clustering and pseudo-label fine-tuning. In addition, we draw from continual learning literature and use learning-without-forgetting to prevent catastrophic forgetting across those training stages. Finally, we show how this deep-clustering process can become part of an incremental discovery strategy with human-in-the-loop. We report results on both CLINC and BANKING datasets. CDI outperforms previous works by a significant margin: 10.26% and 11.72% respectively.
Authors: Ariel Neufeld, Philipp Schmocker
Abstract: In this paper, we generalize the universal approximation property of single-hidden-layer feed-forward neural networks beyond the classical formulation over compact domains. More precisely, by assuming that the activation function is non-polynomial, we derive universal approximation results for neural networks within function spaces over non-compact subsets of a Euclidean space, e.g., weighted spaces, $L^p$-spaces, and (weighted) Sobolev spaces over unbounded domains, where the latter includes the approximation of the (weak) derivatives. Furthermore, we provide some dimension-independent rates for approximating a function with sufficiently regular and integrable Fourier transform by neural networks with non-polynomial activation function.
Authors: Vasileios Vatellis
Abstract: In an era increasingly focused on green computing and explainable AI, revisiting traditional approaches in theoretical and phenomenological particle physics is paramount. This project evaluates various machine learning (ML) algorithms-including Nearest Neighbors, Decision Trees, Random Forest, AdaBoost, Naive Bayes, Quadratic Discriminant Analysis (QDA), and XGBoost-alongside standard neural networks and a novel Physics-Informed Neural Network (PINN) for physics data analysis. We apply these techniques to a binary classification task that distinguishes the experimental viability of simulated scenarios based on Higgs observables and essential parameters. Through this comprehensive analysis, we aim to showcase the capabilities and computational efficiency of each model in binary classification tasks, thereby contributing to the ongoing discourse on integrating ML and Deep Neural Networks (DNNs) into physics research. In this study, XGBoost emerged as the preferred choice among the evaluated machine learning algorithms for its speed and effectiveness, especially in the initial stages of computation with limited datasets. However, while standard Neural Networks and Physics-Informed Neural Networks (PINNs) demonstrated superior performance in terms of accuracy and adherence to physical laws, they require more computational time. These findings underscore the trade-offs between computational efficiency and model sophistication.
Authors: Enkhbold Nyamsuren
Abstract: Democratization of AI is an important topic within the broader topic of the digital divide. This issue is relevant to LLMs, which are becoming popular as AI co-pilots but suffer from a lack of accessibility due to high computational demand. In this study, we evaluate whether quantization is a viable approach toward enabling LLMs on generic consumer devices. The study assesses the performance of five quantized code LLMs in Lua code generation tasks. To evaluate the impact of quantization, the models with 7B parameters were tested on a consumer laptop at 2-, 4-, and 8-bit integer precisions and compared to non-quantized code LLMs with 1.3, 2, and 3 billion parameters. Lua is chosen as a low-level resource language to avoid models' biases related to high-resource languages. The results suggest that the models quantized at the 4-bit integer precision offer the best trade-off between performance and model size. These models can be comfortably deployed on an average laptop without a dedicated GPU. The performance significantly drops at the 2-bit integer precision. The models at 8-bit integer precision require more inference time that does not effectively translate to better performance. The 4-bit models with 7 billion parameters also considerably outperform non-quantized models with lower parameter numbers despite having comparable model sizes with respect to storage and memory demand. While quantization indeed increases the accessibility of smaller LLMs with 7 billion parameters, these LLMs demonstrate overall low performance (less than 50\%) on high-precision and low-resource tasks such as Lua code generation. While accessibility is improved, usability is still not at the practical level comparable to foundational LLMs such as GPT-4o or Llama 3.1 405B.
Authors: Mengqi Wang, Krishna Kumar, Y. T. Feng, Tongming Qu, Min Wang
Abstract: Artificial intelligence (AI) has become a buzz word since Google's AlphaGo beat a world champion in 2017. In the past five years, machine learning as a subset of the broader category of AI has obtained considerable attention in the research community of granular materials. This work offers a detailed review of the recent advances in machine learning-aided studies of granular materials from the particle-particle interaction at the grain level to the macroscopic simulations of granular flow. This work will start with the application of machine learning in the microscopic particle-particle interaction and associated contact models. Then, different neural networks for learning the constitutive behaviour of granular materials will be reviewed and compared. Finally, the macroscopic simulations of practical engineering or boundary value problems based on the combination of neural networks and numerical methods are discussed. We hope readers will have a clear idea of the development of machine learning-aided modelling of granular materials via this comprehensive review work.
Authors: Elynn Chen, Yuefeng Han, Jiayu Li
Abstract: Tensor classification has gained prominence across various fields, yet the challenge of handling partially observed tensor data in real-world applications remains largely unaddressed. This paper introduces a novel approach to tensor classification with incomplete data, framed within the tensor high-dimensional linear discriminant analysis. Specifically, we consider a high-dimensional tensor predictor with missing observations under the Missing Completely at Random (MCR) assumption and employ the Tensor Gaussian Mixture Model to capture the relationship between the tensor predictor and class label. We propose the Tensor LDA-MD algorithm, which manages high-dimensional tensor predictors with missing entries by leveraging the low-rank structure of the discriminant tensor. A key feature of our approach is a novel covariance estimation method under the tensor-based MCR model, supported by theoretical results that allow for correlated entries under mild conditions. Our work establishes the convergence rate of the estimation error of the discriminant tensor with incomplete data and minimax optimal bounds for the misclassification rate, addressing key gaps in the literature. Additionally, we derive large deviation results for the generalized mode-wise (separable) sample covariance matrix and its inverse, which are crucial tools in our analysis and hold independent interest. Our method demonstrates excellent performance in simulations and real data analysis, even with significant proportions of missing data. This research advances high-dimensional LDA and tensor learning, providing practical tools for applications with incomplete data and a solid theoretical foundation for classification accuracy in complex settings.
Authors: Simone Bombari, Marco Mondelli
Abstract: Differentially private gradient descent (DP-GD) is a popular algorithm to train deep learning models with provable guarantees on the privacy of the training data. In the last decade, the problem of understanding its performance cost with respect to standard GD has received remarkable attention from the research community, which formally derived upper bounds on the excess population risk $R_{P}$ in different learning settings. However, existing bounds typically degrade with over-parameterization, i.e., as the number of parameters $p$ gets larger than the number of training samples $n$ -- a regime which is ubiquitous in current deep-learning practice. As a result, the lack of theoretical insights leaves practitioners without clear guidance, leading some to reduce the effective number of trainable parameters to improve performance, while others use larger models to achieve better results through scale. In this work, we show that in the popular random features model with quadratic loss, for any sufficiently large $p$, privacy can be obtained for free, i.e., $\left|R_{P} \right| = o(1)$, not only when the privacy parameter $\varepsilon$ has constant order, but also in the strongly private setting $\varepsilon = o(1)$. This challenges the common wisdom that over-parameterization inherently hinders performance in private learning.
Authors: Takashi Furuya, Anastasis Kratsios
Abstract: Forward-backwards stochastic differential equations (FBSDEs) are central in optimal control, game theory, economics, and mathematical finance. Unfortunately, the available FBSDE solvers operate on \textit{individual} FBSDEs, meaning that they cannot provide a computationally feasible strategy for solving large families of FBSDEs as these solvers must be re-run several times. \textit{Neural operators} (NOs) offer an alternative approach for \textit{simultaneously solving} large families of FBSDEs by directly approximating the solution operator mapping \textit{inputs:} terminal conditions and dynamics of the backwards process to \textit{outputs:} solutions to the associated FBSDE. Though universal approximation theorems (UATs) guarantee the existence of such NOs, these NOs are unrealistically large. We confirm that ``small'' NOs can uniformly approximate the solution operator to structured families of FBSDEs with random terminal time, uniformly on suitable compact sets determined by Sobolev norms, to any prescribed error $\varepsilon>0$ using a depth of $\mathcal{O}(\log(1/\varepsilon))$, a width of $\mathcal{O}(1)$, and a sub-linear rank; i.e. $\mathcal{O}(1/\varepsilon^r)$ for some $r<1$. This result is rooted in our second main contribution, which shows that convolutional NOs of similar depth, width, and rank can approximate the solution operator to a broad class of Elliptic PDEs. A key insight here is that the convolutional layers of our NO can efficiently encode the Green's function associated to the Elliptic PDEs linked to our FBSDEs. A byproduct of our analysis is the first theoretical justification for the benefit of lifting channels in NOs: they exponentially decelerate the growth rate of the NO's rank.
Authors: Yuki Ohnishi, Jordan Awan
Abstract: Differential privacy is the leading mathematical framework for privacy protection, providing a probabilistic guarantee that safeguards individuals' private information when publishing statistics from a dataset. This guarantee is achieved by applying a randomized algorithm to the original data, which introduces unique challenges in data analysis by distorting inherent patterns. In particular, causal inference using observational data in privacy-sensitive contexts is challenging because it requires covariate balance between treatment groups, yet checking the true covariates is prohibited to prevent leakage of sensitive information. In this article, we present a differentially private two-stage covariate balancing weighting estimator to infer causal effects from observational data. Our algorithm produces both point and interval estimators with statistical guarantees, such as consistency and rate optimality, under a given privacy budget.
Authors: Steven Triplett, Simon Minami, Rakesh Verma
Abstract: In the modern age an enormous amount of communication occurs online, and it is difficult to know when something written is genuine or deceitful. There are many reasons for someone to deceive online (e.g., monetary gain, political gain) and detecting this behavior without any physical interaction is a difficult task. Additionally, deception occurs in several text-only domains and it is unclear if these various sources can be leveraged to improve detection. To address this, eight datasets were utilized from various domains to evaluate their effect on classifier performance when combined with transfer learning via intermediate layer concatenation of fine-tuned BERT models. We find improvements in accuracy over the baseline. Furthermore, we evaluate multiple distance measurements between datasets and find that Jensen-Shannon distance correlates moderately with transfer learning performance. Finally, the impact was evaluated of multiple methods, which produce additional information in a dataset's text via named entities, on BERT performance and we find notable improvement in accuracy of up to 11.2%.
Authors: Raviraj Joshi, Kanishk Singla, Anusha Kamath, Raunak Kalani, Rakesh Paul, Utkarsh Vaidya, Sanjay Singh Chauhan, Niranjan Wartikar, Eileen Long
Abstract: Multilingual LLMs support a variety of languages; however, their performance is suboptimal for low-resource languages. In this work, we emphasize the importance of continued pre-training of multilingual LLMs and the use of translation-based synthetic pre-training corpora for improving LLMs in low-resource languages. We conduct our study in the context of the low-resource Indic language Hindi. We introduce Nemotron-Mini-Hindi 4B, a bilingual SLM supporting both Hindi and English, based on Nemotron-Mini 4B. The model is trained using a mix of real and synthetic Hindi + English tokens, with continuous pre-training performed on 400B tokens. We demonstrate that both the base and instruct models achieve state-of-the-art results on Hindi benchmarks while remaining competitive on English tasks. Additionally, we observe that the continued pre-training approach enhances the model's overall factual accuracy.
Authors: Eric Elmoznino, Thomas Jiralerspong, Yoshua Bengio, Guillaume Lajoie
Abstract: Compositionality is believed to be fundamental to intelligence. In humans, it underlies the structure of thought, language, and higher-level reasoning. In AI, compositional representations can enable a powerful form of out-of-distribution generalization, in which a model systematically adapts to novel combinations of known concepts. However, while we have strong intuitions about what compositionality is, there currently exists no formal definition for it that is measurable and mathematical. Here, we propose such a definition, which we call representational compositionality, that accounts for and extends our intuitions about compositionality. The definition is conceptually simple, quantitative, grounded in algorithmic information theory, and applicable to any representation. Intuitively, representational compositionality states that a compositional representation satisfies three properties. First, it must be expressive. Second, it must be possible to re-describe the representation as a function of discrete symbolic sequences with re-combinable parts, analogous to sentences in natural language. Third, the function that relates these symbolic sequences to the representation, analogous to semantics in natural language, must be simple. Through experiments on both synthetic and real world data, we validate our definition of compositionality and show how it unifies disparate intuitions from across the literature in both AI and cognitive science. We also show that representational compositionality, while theoretically intractable, can be readily estimated using standard deep learning tools. Our definition has the potential to inspire the design of novel, theoretically-driven models that better capture the mechanisms of compositional thought.
Authors: Lechen Zhang, Tolga Ergen, Lajanugen Logeswaran, Moontae Lee, David Jurgens
Abstract: Large Language Models (LLMs) have shown impressive capabilities in many scenarios, but their performance depends, in part, on the choice of prompt. Past research has focused on optimizing prompts specific to a task. However, much less attention has been given to optimizing the general instructions included in a prompt, known as a system prompt. To address this gap, we propose SPRIG, an edit-based genetic algorithm that iteratively constructs prompts from prespecified components to maximize the model's performance in general scenarios. We evaluate the performance of system prompts on a collection of 47 different types of tasks to ensure generalizability. Our study finds that a single optimized system prompt performs on par with task prompts optimized for each individual task. Moreover, combining system and task-level optimizations leads to further improvement, which showcases their complementary nature. Experiments also reveal that the optimized system prompts generalize effectively across model families, parameter sizes, and languages. This study provides insights into the role of system-level instructions in maximizing LLM potential.
Authors: Zedian Shao, Hongbin Liu, Jaden Mu, Neil Zhenqiang Gong
Abstract: In a prompt injection attack, an attacker injects a prompt into the original one, aiming to make the LLM follow the injected prompt and perform a task chosen by the attacker. Existing prompt injection attacks primarily focus on how to blend the injected prompt into the original prompt without altering the LLM itself. Our experiments show that these attacks achieve some success, but there is still significant room for improvement. In this work, we show that an attacker can boost the success of prompt injection attacks by poisoning the LLM's alignment process. Specifically, we propose PoisonedAlign, a method to strategically create poisoned alignment samples. When even a small fraction of the alignment data is poisoned using our method, the aligned LLM becomes more vulnerable to prompt injection while maintaining its foundational capabilities. The code is available at https://github.com/Sadcardation/PoisonedAlign
Authors: Arpan Mahara, Md Rezaul Karim Khan, Naphtali D. Rishe, Wenjia Wang, Seyed Masoud Sadjadi
Abstract: Road Extraction is a sub-domain of Remote Sensing applications; it is a subject of extensive and ongoing research. The procedure of automatically extracting roads from satellite imagery encounters significant challenges due to the multi-scale and diverse structures of roads; improvement in this field is needed. The DeepLab series, known for its proficiency in semantic segmentation due to its efficiency in interpreting multi-scale objects' features, addresses some of these challenges caused by the varying nature of roads. The present work proposes the utilization of DeepLabV3+, the latest version of the DeepLab series, by introducing an innovative Dense Depthwise Dilated Separable Spatial Pyramid Pooling (DenseDDSSPP) module and integrating it in place of the conventional Atrous Spatial Pyramid Pooling (ASPP) module. This modification enhances the extraction of complex road structures from satellite images. This study hypothesizes that the integration of DenseDDSSPP, combined with an appropriately selected backbone network and a Squeeze-and-Excitation block, will generate an efficient dense feature map by focusing on relevant features, leading to more precise and accurate road extraction from Remote Sensing images. The results section presents a comparison of our model's performance against state-of-the-art models, demonstrating better results that highlight the effectiveness and success of the proposed approach.
Authors: Adel Javanmard, Jingwei Ji, Renyuan Xu
Abstract: We study the dynamic pricing problem faced by a broker that buys and sells a large number of financial securities in the credit market, such as corporate bonds, government bonds, loans, and other credit-related securities. One challenge in pricing these securities is their infrequent trading, which leads to insufficient data for individual pricing. However, many of these securities share structural features that can be utilized. Building on this, we propose a multi-task dynamic pricing framework that leverages these shared structures across securities, enhancing pricing accuracy through learning. In our framework, a security is fully characterized by a $d$ dimensional contextual/feature vector. The customer will buy (sell) the security from the broker if the broker quotes a price lower (higher) than that of the competitors. We assume a linear contextual model for the competitor's pricing, with unknown parameters a prior. The parameters for pricing different securities may or may not be similar to each other. The firm's objective is to minimize the expected regret, namely, the expected revenue loss against a clairvoyant policy which has the knowledge of the parameters of the competitor's pricing model. We show that the regret of our policy is better than both a policy that treats each security individually and a policy that treats all securities as the same. Moreover, the regret is bounded by $\tilde{O} ( \delta_{\max} \sqrt{T M d} + M d ) $, where $M$ is the number of securities and $\delta_{\max}$ characterizes the overall dissimilarity across securities in the basket.
Authors: Jinlin Lai, Yuling Yao
Abstract: Vanilla variational inference finds an optimal approximation to the Bayesian posterior distribution, but even the exact Bayesian posterior is often not meaningful under model misspecification. We propose predictive variational inference (PVI): a general inference framework that seeks and samples from an optimal posterior density such that the resulting posterior predictive distribution is as close to the true data generating process as possible, while this this closeness is measured by multiple scoring rules. By optimizing the objective, the predictive variational inference is generally not the same as, or even attempting to approximate, the Bayesian posterior, even asymptotically. Rather, we interpret it as implicit hierarchical expansion. Further, the learned posterior uncertainty detects heterogeneity of parameters among the population, enabling automatic model diagnosis. This framework applies to both likelihood-exact and likelihood-free models. We demonstrate its application in real data examples.
Authors: Shantanu Thorat, Tianbao Yang
Abstract: As LLMs increase in accessibility, LLM-generated texts have proliferated across several fields, such as scientific, academic, and creative writing. However, LLMs are not created equally; they may have different architectures and training datasets. Thus, some LLMs may be more challenging to detect than others. Using two datasets spanning four total writing domains, we train AI-generated (AIG) text classifiers using the LibAUC library - a deep learning library for training classifiers with imbalanced datasets. Our results in the Deepfake Text dataset show that AIG-text detection varies across domains, with scientific writing being relatively challenging. In the Rewritten Ivy Panda (RIP) dataset focusing on student essays, we find that the OpenAI family of LLMs was substantially difficult for our classifiers to distinguish from human texts. Additionally, we explore possible factors that could explain the difficulties in detecting OpenAI-generated texts.
Authors: Zelei Cheng, Xian Wu, Jiahao Yu, Shuo Han, Xin-Qiang Cai, Xinyu Xing
Abstract: Toxicity classification in textual content remains a significant problem. Data with labels from a single annotator fall short of capturing the diversity of human perspectives. Therefore, there is a growing need to incorporate crowdsourced annotations for training an effective toxicity classifier. Additionally, the standard approach to training a classifier using empirical risk minimization (ERM) may fail to address the potential shifts between the training set and testing set due to exploiting spurious correlations. This work introduces a novel bi-level optimization framework that integrates crowdsourced annotations with the soft-labeling technique and optimizes the soft-label weights by Group Distributionally Robust Optimization (GroupDRO) to enhance the robustness against out-of-distribution (OOD) risk. We theoretically prove the convergence of our bi-level optimization algorithm. Experimental results demonstrate that our approach outperforms existing baseline methods in terms of both average and worst-group accuracy, confirming its effectiveness in leveraging crowdsourced annotations to achieve more effective and robust toxicity classification.
Authors: Mingyuan Zhou, Huangjie Zheng, Yi Gu, Zhendong Wang, Hai Huang
Abstract: Score identity Distillation (SiD) is a data-free method that has achieved state-of-the-art performance in image generation by leveraging only a pretrained diffusion model, without requiring any training data. However, the ultimate performance of SiD is constrained by the accuracy with which the pretrained model captures the true data scores at different stages of the diffusion process. In this paper, we introduce SiDA (SiD with Adversarial Loss), which not only enhances generation quality but also improves distillation efficiency by incorporating real images and adversarial loss. SiDA utilizes the encoder from the generator's score network as a discriminator, boosting its ability to distinguish between real images and those generated by SiD. The adversarial loss is batch-normalized within each GPU and then combined with the original SiD loss. This integration effectively incorporates the average "fakeness" per GPU batch into the pixel-based SiD loss, enabling SiDA to distill a single-step generator either from scratch or by fine-tuning an existing one. SiDA converges significantly faster than its predecessor when trained from scratch, and swiftly improves upon the original model's performance after an initial warmup period during fine-tuning from a pre-distilled SiD generator. This one-step adversarial distillation method has set new benchmarks for generation performance when distilling EDM diffusion models pretrained on CIFAR-10 (32x32) and ImageNet (64x64), achieving FID scores of $\mathbf{1.499}$ on CIFAR-10 unconditional, $\mathbf{1.396}$ on CIFAR-10 conditional, and $\mathbf{1.110}$ on ImageNet 64x64. Our open-source code will be integrated into the SiD codebase on GitHub.
Authors: Zijie Zhao, Roy E. Welsch
Abstract: Leveraging Deep Reinforcement Learning (DRL) in automated stock trading has shown promising results, yet its application faces significant challenges, including the curse of dimensionality, inertia in trading actions, and insufficient portfolio diversification. Addressing these challenges, we introduce the Hierarchical Reinforced Trader (HRT), a novel trading strategy employing a bi-level Hierarchical Reinforcement Learning framework. The HRT integrates a Proximal Policy Optimization (PPO)-based High-Level Controller (HLC) for strategic stock selection with a Deep Deterministic Policy Gradient (DDPG)-based Low-Level Controller (LLC) tasked with optimizing trade executions to enhance portfolio value. In our empirical analysis, comparing the HRT agent with standalone DRL models and the S&P 500 benchmark during both bullish and bearish market conditions, we achieve a positive and higher Sharpe ratio. This advancement not only underscores the efficacy of incorporating hierarchical structures into DRL strategies but also mitigates the aforementioned challenges, paving the way for designing more profitable and robust trading algorithms in complex markets.
Authors: Y. Qiang Sun, Pedram Hassanzadeh, Mohsen Zand, Ashesh Chattopadhyay, Jonathan Weare, Dorian S. Abbot
Abstract: Predicting gray swan weather extremes, which are possible but so rare that they are absent from the training dataset, is a major concern for AI weather/climate models. An important open question is whether AI models can extrapolate from weaker weather events present in the training set to stronger, unseen weather extremes. To test this, we train independent versions of the AI model FourCastNet on the 1979-2015 ERA5 dataset with all data, or with Category 3-5 tropical cyclones (TCs) removed, either globally or only over the North Atlantic or Western Pacific basin. We then test these versions of FourCastNet on 2018-2023 Category 5 TCs (gray swans). All versions yield similar accuracy for global weather, but the one trained without Category 3-5 TCs cannot accurately forecast Category 5 TCs, indicating that these models cannot extrapolate from weaker storms. The versions trained without Category 3-5 TCs in one basin show some skill forecasting Category 5 TCs in that basin, suggesting that FourCastNet can generalize across tropical basins. This is encouraging and surprising because regional information is implicitly encoded in inputs. No version satisfies gradient-wind balance, implying that enforcing such physical constraints may not improve generalizability to gray swans. Given that current state-of-the-art AI weather/climate models have similar learning strategies, we expect our findings to apply to other models and extreme events. Our work demonstrates that novel learning strategies are needed for AI weather/climate models to provide early warning or estimated statistics for the rarest, most impactful weather extremes.
Authors: Francisco M\'arcio Barboza, Arthur Anthony da Cunha Rom\~ao E Silva, Bruno Motta de Carvalho
Abstract: Basement relief gravimetry is crucial in geophysics, especially for oil exploration and mineral prospecting. It involves solving an inverse problem to infer geological model parameters from observed data. The model represents basement relief with constant-density prisms, and the data reflect gravitational anomalies from these prisms. Inverse problems are often ill-posed, meaning small data changes can lead to large solution variations. To mitigate this, regularization techniques like Tikhonov's are used to stabilize solutions. This study compares regularization methods applied to gravimetric inversion, including Smoothness Constraints, Total Variation, Discrete Cosine Transform (DCT), and Discrete Wavelet Transform (DWT) using Daubechies D4 wavelets. Optimization, particularly with Genetic Algorithms (GA), is used to find prism depths that best match observed anomalies. GA, inspired by natural selection, selects the best solutions to minimize the objective function. The results, evaluated through fit metrics and error analysis, show the effectiveness of all regularization methods and GA, with the Smoothness constraint performing best in synthetic models. For the real data model, all methods performed similarly.
Authors: Mojtaba Heydari, Mehrez Souden, Bruno Conejo, Joshua Atkins
Abstract: We introduce ImmerseDiffusion, an end-to-end generative audio model that produces 3D immersive soundscapes conditioned on the spatial, temporal, and environmental conditions of sound objects. ImmerseDiffusion is trained to generate first-order ambisonics (FOA) audio, which is a conventional spatial audio format comprising four channels that can be rendered to multichannel spatial output. The proposed generative system is composed of a spatial audio codec that maps FOA audio to latent components, a latent diffusion model trained based on various user input types, namely, text prompts, spatial, temporal and environmental acoustic parameters, and optionally a spatial audio and text encoder trained in a Contrastive Language and Audio Pretraining (CLAP) style. We propose metrics to evaluate the quality and spatial adherence of the generated spatial audio. Finally, we assess the model performance in terms of generation quality and spatial conformance, comparing the two proposed modes: ``descriptive", which uses spatial text prompts) and ``parametric", which uses non-spatial text prompts and spatial parameters. Our evaluations demonstrate promising results that are consistent with the user conditions and reflect reliable spatial fidelity.
Authors: Bryan Chan, Anson Leung, James Bergstra
Abstract: Offline-to-online reinforcement learning (O2O RL) aims to obtain a continually improving policy as it interacts with the environment, while ensuring the initial behaviour is satisficing. This satisficing behaviour is necessary for robotic manipulation where random exploration can be costly due to catastrophic failures and time. O2O RL is especially compelling when we can only obtain a scarce amount of (potentially suboptimal) demonstrations$\unicode{x2014}$a scenario where behavioural cloning (BC) is known to suffer from distribution shift. Previous works have outlined the challenges in applying O2O RL algorithms under the image-based environments. In this work, we propose a novel O2O RL algorithm that can learn in a real-life image-based robotic vacuum grasping task with a small number of demonstrations where BC fails majority of the time. The proposed algorithm replaces the target network in off-policy actor-critic algorithms with a regularization technique inspired by neural tangent kernel. We demonstrate that the proposed algorithm can reach above 90% success rate in under two hours of interaction time, with only 50 human demonstrations, while BC and two commonly-used RL algorithms fail to achieve similar performance.
Authors: Kazuhiro Ohta, Satoshi Ono
Abstract: Neural Radiance Field (NeRF), capable of synthesizing high-quality novel viewpoint images, suffers from issues like artifact occurrence due to its fixed sampling points during rendering. This study proposes a method that optimizes sampling points to reduce artifacts and produce more detailed images.
Authors: Ryan Diaz, Adam Imdieke, Vivek Veeriah, Karthik Desingh
Abstract: This paper primarily focuses on learning robust visual-force policies in the context of high-precision object assembly tasks. Specifically, we focus on the contact phase of the assembly task where both objects (peg and hole) have made contact and the objective lies in maneuvering the objects to complete the assembly. Moreover, we aim to learn contact-rich manipulation policies with multisensory inputs on limited expert data by expanding human demonstrations via online data augmentation. We develop a simulation environment with a dual-arm robot manipulator to evaluate the effect of augmented expert demonstration data. Our focus is on evaluating the robustness of our model with respect to certain task variations: grasp pose, peg/hole shape, object body shape, scene appearance, camera pose, and force-torque/proprioception noise. We show that our proposed data augmentation method helps in learning a multisensory manipulation policy that is robust to unseen instances of these variations, particularly physical variations such as grasp pose. Additionally, our ablative studies show the significant contribution of force-torque data to the robustness of our model. For additional experiments and qualitative results, we refer to the project webpage at https://bit.ly/47skWXH .
URLs: https://bit.ly/47skWXH
Authors: Marie Roald, Magnus Breder Birkenes, Lars Gunnars{\o}nn Bag{\o}ien Johnsen
Abstract: Digital tools for text analysis have long been essential for the searchability and accessibility of digitised library collections. Recent computer vision advances have introduced similar capabilities for visual materials, with deep learning-based embeddings showing promise for analysing visual heritage. Given that many books feature visuals in addition to text, taking advantage of these breakthroughs is critical to making library collections open and accessible. In this work, we present a proof-of-concept image search application for exploring images in the National Library of Norway's pre-1900 books, comparing Vision Transformer (ViT), Contrastive Language-Image Pre-training (CLIP), and Sigmoid loss for Language-Image Pre-training (SigLIP) embeddings for image retrieval and classification. Our results show that the application performs well for exact image retrieval, with SigLIP embeddings slightly outperforming CLIP and ViT in both retrieval and classification tasks. Additionally, SigLIP-based image classification can aid in cleaning image datasets from a digitisation pipeline.
Authors: Xiaohang Xu, Renhe Jiang, Chuang Yang, Zipei Fan, Kaoru Sezaki
Abstract: With the popularity of location-based services, human mobility prediction plays a key role in enhancing personalized navigation, optimizing recommendation systems, and facilitating urban mobility and planning. This involves predicting a user's next POI (point-of-interest) visit using their past visit history. However, the uneven distribution of visitations over time and space, namely the long-tail problem in spatial distribution, makes it difficult for AI models to predict those POIs that are less visited by humans. In light of this issue, we propose the Long-Tail Adjusted Next POI Prediction (LoTNext) framework for mobility prediction, combining a Long-Tailed Graph Adjustment module to reduce the impact of the long-tailed nodes in the user-POI interaction graph and a novel Long-Tailed Loss Adjustment module to adjust loss by logit score and sample weight adjustment strategy. Also, we employ the auxiliary prediction task to enhance generalization and accuracy. Our experiments with two real-world trajectory datasets demonstrate that LoTNext significantly surpasses existing state-of-the-art works. Our code is available at https://github.com/Yukayo/LoTNext.
Authors: Suning Huang, Zheyu Zhang, Tianhai Liang, Yihan Xu, Zhehao Kou, Chenhao Lu, Guowei Xu, Zhengrong Xue, Huazhe Xu
Abstract: Visual deep reinforcement learning (RL) enables robots to acquire skills from visual input for unstructured tasks. However, current algorithms suffer from low sample efficiency, limiting their practical applicability. In this work, we present MENTOR, a method that improves both the architecture and optimization of RL agents. Specifically, MENTOR replaces the standard multi-layer perceptron (MLP) with a mixture-of-experts (MoE) backbone, enhancing the agent's ability to handle complex tasks by leveraging modular expert learning to avoid gradient conflicts. Furthermore, MENTOR introduces a task-oriented perturbation mechanism, which heuristically samples perturbation candidates containing task-relevant information, leading to more targeted and effective optimization. MENTOR outperforms state-of-the-art methods across three simulation domains -- DeepMind Control Suite, Meta-World, and Adroit. Additionally, MENTOR achieves an average of 83% success rate on three challenging real-world robotic manipulation tasks including peg insertion, cable routing, and tabletop golf, which significantly surpasses the success rate of 32% from the current strongest model-free visual RL algorithm. These results underscore the importance of sample efficiency in advancing visual RL for real-world robotics. Experimental videos are available at https://suninghuang19.github.io/mentor_page.
Authors: Wei Xie, Shuoyoucheng Ma, Zhenhua Wang, Enze Wang, Baosheng Wang, Jinshu Su
Abstract: Despite their proficiency in math tasks, the mechanisms underlying LLMs' mathematical reasoning abilities remain a subject of debate. Recent studies suggest that chain-of-thought (CoT) prompts can bolster mathematical reasoning by encouraging LLMs to employ human-like logical reasoning (System 2), enabling them to excel on the Cognitive Reflection Test (CRT). To assess whether LLMs genuinely possess System 2-like logical reasoning, we introduced targeted modifications to CRT problems. Our findings reveal that, despite the use of CoT prompts, mainstream LLMs, including the latest o1-preview model, continue to exhibit a significant error rate. Further analysis indicates that they predominantly rely on System 1-like intuitive reasoning and pattern matching derived from training data, rather than demonstrating mastery of mathematical thinking. This discovery challenges the prevailing notion that LLMs possess genuine logical reasoning abilities and that CoT can enhance them. Consequently, this work may temper overly optimistic projections regarding LLMs' advancement toward artificial general intelligence.
Authors: Noam Ginio (Faculty of Civil and Environmental Engineering, Technion, Haifa, Israel), Michael Lindenbaum (Faculty of Computer Science, Technion, Haifa, Israel, Interdisciplinary program for Marine Engineering, Technion, Haifa, Israel), Barak Fishbain (Faculty of Civil and Environmental Engineering, Technion, Haifa, Israel), Dan Liberzon (Faculty of Civil and Environmental Engineering, Technion, Haifa, Israel, Interdisciplinary program for Marine Engineering, Technion, Haifa, Israel)
Abstract: Effective spatio-temporal measurements of water surface elevation (water waves) in laboratory experiments are crucial for scientific and engineering research. Existing techniques are often cumbersome, computationally heavy and generally suffer from limitations in wavenumber/frequency response. To address these challenges, we propose Wave (from) Polarized Light Learning (WPLL), a learning based remote sensing method for laboratory implementation, capable of inferring surface elevation and slope maps in high resolution. The method uses the polarization properties of light reflected from the water surface. The WPLL uses a deep neural network (DNN) model that approximates the water surface slopes from the polarized light intensities. Once trained on simple monochromatic wave trains, the WPLL is capable of producing high-resolution and accurate 2D reconstruction of the water surface slopes and elevation in a variety of irregular wave fields. The method's robustness is demonstrated by showcasing its high wavenumber/frequency response, its ability to reconstruct wave fields propagating at arbitrary angles relative to the camera optical axis, and its computational efficiency. This developed methodology is an accurate and cost-effective near-real time remote sensing tool for laboratory water surface waves measurements, setting the path for upscaling to open sea application for research, monitoring, and short-time forecasting.
Authors: Sivangi Chatterjee, Srishti Ganguly, Avik Bose, Hrithik Raj Prasad, Arijit Ghosal
Abstract: This project explores the application of machine learning techniques for music genre classification using the GTZAN dataset, which contains 100 audio files per genre. Motivated by the growing demand for personalized music recommendations, we focused on classifying five genres-Blues, Classical, Jazz, Hip Hop, and Country-using a variety of algorithms including Logistic Regression, K-Nearest Neighbors (KNN), Random Forest, and Artificial Neural Networks (ANN) implemented via Keras. The ANN model demonstrated the best performance, achieving a validation accuracy of 92.44%. We also analyzed key audio features such as spectral roll-off, spectral centroid, and MFCCs, which helped enhance the model's accuracy. Future work will expand the model to cover all ten genres, investigate advanced methods like Long Short-Term Memory (LSTM) networks and ensemble approaches, and develop a web application for real-time genre classification and playlist generation. This research aims to contribute to improving music recommendation systems and content curation.
Authors: Chen Yan, Weina Wang, Lei Ying
Abstract: We study the finite horizon Restless Multi-Armed Bandit (RMAB) problem with $N$ homogeneous arms, focusing on the challenges posed by degenerate RMABs, which are prevalent in practical applications. While previous work has shown that Linear Programming (LP)-based policies achieve exponentially fast convergence relative to the LP upper bound in non-degenerate models, applying these LP-based policies to degenerate RMABs results in slower convergence rates of $O(1/\sqrt{N})$. We construct a diffusion system that incorporates both the mean and variance of the stochastic processes, in contrast to the fluid system from the LP, which only accounts for the mean, thereby providing a more accurate representation of RMAB dynamics. Consequently, our novel diffusion-resolving policy achieves an optimality gap of $O(1/N)$ relative to the true optimal value, rather than the LP upper bound, revealing that the fluid approximation and the LP upper bound are too loose in degenerate settings. These insights pave the way for constructing policies that surpass the $O(1/\sqrt{N})$ optimality gap for any RMAB, whether degenerate or not.
Authors: M. Rostami, S. S. Kia
Abstract: In this article, we consider the problem of unconstrained time-varying convex optimization, where the cost function changes with time. We provide an in-depth technical analysis of the problem and argue why freezing the cost at each time step and taking finite steps toward the minimizer is not the best tracking solution for this problem. We propose a set of algorithms that by taking into account the temporal variation of the cost aim to reduce the tracking error of the time-varying minimizer of the problem. The main contribution of our work is that our proposed algorithms only require the first-order derivatives of the cost function with respect to the decision variable. This approach significantly reduces computational cost compared to the existing algorithms, which use the inverse of the Hessian of the cost. Specifically, the proposed algorithms reduce the computational cost from $O(n^3)$ to $O(n)$ per timestep, where $n$ is the size of the decision variable. Avoiding the inverse of the Hessian also makes our algorithms applicable to non-convex optimization problems. We refer to these algorithms as $O(n)$-algorithms. These $O(n)$-algorithms are designed to solve the problem for different scenarios based on the available temporal information about the cost. We illustrate our results through various examples, including the solution of a model predictive control problem framed as a convex optimization problem with a streaming time-varying cost function.
Authors: Jiahao Wang, Amer Shalaby
Abstract: Users of the transit system flood social networks daily with messages that contain valuable insights crucial for improving service quality. These posts help transit agencies quickly identify emerging issues. Parsing topics and sentiments is key to gaining comprehensive insights to foster service excellence. However, the volume of messages makes manual analysis impractical, and standard NLP techniques like Term Frequency-Inverse Document Frequency (TF-IDF) fall short in nuanced interpretation. Traditional sentiment analysis separates topics and sentiments before integrating them, often missing the interaction between them. This incremental approach complicates classification and reduces analytical productivity. To address these challenges, we propose a novel approach to extracting and analyzing transit-related information, including sentiment and sarcasm detection, identification of unusual system problems, and location data from social media. Our method employs Large Language Models (LLM), specifically Llama 3, for a streamlined analysis free from pre-established topic labels. To enhance the model's domain-specific knowledge, we utilize Retrieval-Augmented Generation (RAG), integrating external knowledge sources into the information extraction pipeline. We validated our method through extensive experiments comparing its performance with traditional NLP approaches on user tweet data from the real world transit system. Our results demonstrate the potential of LLMs to transform social media data analysis in the public transit domain, providing actionable insights and enhancing transit agencies' responsiveness by extracting a broader range of information.
Authors: Nguyen Thang Loi, Duong Tan Loc, Vo Nguyen Le Duy
Abstract: Feature Selection (FS) under domain adaptation (DA) is a critical task in machine learning, especially when dealing with limited target data. However, existing methods lack the capability to guarantee the reliability of FS under DA. In this paper, we introduce a novel statistical method to statistically test FS reliability under DA, named SFS-DA (statistical FS-DA). The key strength of SFS-DA lies in its ability to control the false positive rate (FPR) below a pre-specified level $\alpha$ (e.g., 0.05) while maximizing the true positive rate. Compared to the literature on statistical FS, SFS-DA presents a unique challenge in addressing the effect of DA to ensure the validity of the inference on FS results. We overcome this challenge by leveraging the Selective Inference (SI) framework. Specifically, by carefully examining the FS process under DA whose operations can be characterized by linear and quadratic inequalities, we prove that achieving FPR control in SFS-DA is indeed possible. Furthermore, we enhance the true detection rate by introducing a more strategic approach. Experiments conducted on both synthetic and real-world datasets robustly support our theoretical results, showcasing the superior performance of the proposed SFS-DA method.
Authors: Vittorio Torri, Elisa Barbieri, Anna Cantarutti, Carlo Giaquinto, Francesca Ieva
Abstract: Objective: Recognizing diseases from discharge letters is crucial for cohort selection and epidemiological analyses, as this is the only type of data consistently produced across hospitals. This is a classic document classification problem, typically requiring supervised learning. However, manual annotation of large datasets of discharge letters is uncommon since it is extremely time-consuming. We propose a novel weakly-supervised pipeline to recognize diseases from Italian discharge letters. Methods: Our Natural Language Processing pipeline is based on a fine-tuned version of the Italian Umberto model. The pipeline extracts diagnosis-related sentences from a subset of letters and applies a two-level clustering using the embeddings generated by the fine-tuned Umberto model. These clusters are summarized and those mapped to the diseases of interest are selected as weak labels. Finally, the same BERT-based model is trained using these weak labels to detect the targeted diseases. Results: A case study related to the identification of bronchiolitis with 33'176 Italian discharge letters from 44 hospitals in the Veneto Region shows the potential of our method, with an AUC of 77.7 % and an F1-Score of 75.1 % on manually annotated labels, improving compared to other non-supervised methods and with a limited loss compared to fully supervised methods. Results are robust to the cluster selection and the identified clusters highlight the potential to recognize a variety of diseases. Conclusions: This study demonstrates the feasibility of diagnosis identification from Italian discharge letters in the absence of labelled data. Our pipeline showed strong performance and robustness, and its flexibility allows for easy adaptation to various diseases. This approach offers a scalable solution for clinical text classification, reducing the need for manual annotation while maintaining good accuracy.
Authors: Chuhan Xie, Kaicheng Jin, Jiadong Liang, Zhihua Zhang
Abstract: We study time-uniform statistical inference for parameters in stochastic approximation (SA), which encompasses a bunch of applications in optimization and machine learning. To that end, we analyze the almost-sure convergence rates of the averaged iterates to a scaled sum of Gaussians in both linear and nonlinear SA problems. We then construct three types of asymptotic confidence sequences that are valid uniformly across all times with coverage guarantees, in an asymptotic sense that the starting time is sufficiently large. These coverage guarantees remain valid if the unknown covariance matrix is replaced by its plug-in estimator, and we conduct experiments to validate our methodology.
Authors: Hrishav Bakul Barua, Stefanov Kalin, Lemuel Lai En Che, Dhall Abhinav, Wong KokSheik, Krishnasamy Ganesh
Abstract: Low Dynamic Range (LDR) to High Dynamic Range (HDR) image translation is an important computer vision problem. There is a significant amount of research utilizing both conventional non-learning methods and modern data-driven approaches, focusing on using both single-exposed and multi-exposed LDR for HDR image reconstruction. However, most current state-of-the-art methods require high-quality paired {LDR,HDR} datasets for model training. In addition, there is limited literature on using unpaired datasets for this task where the model learns a mapping between domains, i.e., LDR to HDR. To address limitations of current methods, such as the paired data constraint , as well as unwanted blurring and visual artifacts in the reconstructed HDR, we propose a method that uses a modified cycle-consistent adversarial architecture and utilizes unpaired {LDR,HDR} datasets for training. The method introduces novel generators to address visual artifact removal and an encoder and loss to address semantic consistency, another under-explored topic. The method achieves state-of-the-art results across several benchmark datasets and reconstructs high-quality HDR images.
Authors: Rajeev Ranjan Dwivedi, Priyadarshini Kumari, Vinod K Kurmi
Abstract: Deep neural networks trained on biased data often inadvertently learn unintended inference rules, particularly when labels are strongly correlated with biased features. Existing bias mitigation methods typically involve either a) predefining bias types and enforcing them as prior knowledge or b) reweighting training samples to emphasize bias-conflicting samples over bias-aligned samples. However, both strategies address bias indirectly in the feature or sample space, with no control over learned weights, making it difficult to control the bias propagation across different layers. Based on this observation, we introduce a novel approach to address bias directly in the model's parameter space, preventing its propagation across layers. Our method involves training two models: a bias model for biased features and a debias model for unbiased details, guided by the bias model. We enforce dissimilarity in the debias model's later layers and similarity in its initial layers with the bias model, ensuring it learns unbiased low-level features without adopting biased high-level abstractions. By incorporating this explicit constraint during training, our approach shows enhanced classification accuracy and debiasing effectiveness across various synthetic and real-world datasets of different sizes. Moreover, the proposed method demonstrates robustness across different bias types and percentages of biased samples in the training data. The code is available at: https://visdomlab.github.io/CosFairNet/
Authors: Yui Lo, Yuqian Chen, Dongnan Liu, Wan Liu, Leo Zekelman, Jarrett Rushmore, Fan Zhang, Yogesh Rathi, Nikos Makris, Alexandra J. Golby, Weidong Cai, Lauren J. O'Donnell
Abstract: The shape of the brain's white matter connections is relatively unexplored in diffusion MRI tractography analysis. While it is known that tract shape varies in populations and across the human lifespan, it is unknown if the variability in dMRI tractography-derived shape may relate to the brain's functional variability across individuals. This work explores the potential of leveraging tractography fiber cluster shape measures to predict subject-specific cognitive performance. We implement machine learning models to predict individual cognitive performance scores. We study a large-scale database from the HCP-YA study. We apply an atlas-based fiber cluster parcellation to the dMRI tractography of each individual. We compute 15 shape, microstructure, and connectivity features for each fiber cluster. Using these features as input, we train a total of 210 models to predict 7 different NIH Toolbox cognitive performance assessments. We apply an explainable AI technique, SHAP, to assess the importance of each fiber cluster for prediction. Our results demonstrate that shape measures are predictive of individual cognitive performance. The studied shape measures, such as irregularity, diameter, total surface area, volume, and branch volume, are as effective for prediction as microstructure and connectivity measures. The overall best-performing feature is a shape feature, irregularity, which describes how different a cluster's shape is from an idealized cylinder. Further interpretation using SHAP values suggest that fiber clusters with features highly predictive of cognitive ability are widespread throughout the brain, including fiber clusters from the superficial association, deep association, cerebellar, striatal, and projection pathways. This study demonstrates the strong potential of shape descriptors to enhance the study of the brain's white matter and its relationship to cognitive function.
Authors: Le Hong Phong, Ho Ngoc Luat, Vo Nguyen Le Duy
Abstract: Detecting the presence of anomalies in regression models is a crucial task in machine learning, as anomalies can significantly impact the accuracy and reliability of predictions. Random Sample Consensus (RANSAC) is one of the most popular robust regression methods for addressing this challenge. However, this method lacks the capability to guarantee the reliability of the anomaly detection (AD) results. In this paper, we propose a novel statistical method for testing the AD results obtained by RANSAC, named CTRL-RANSAC (controllable RANSAC). The key strength of the proposed method lies in its ability to control the probability of misidentifying anomalies below a pre-specified level $\alpha$ (e.g., $\alpha = 0.05$). By examining the selection strategy of RANSAC and leveraging the Selective Inference (SI) framework, we prove that achieving controllable RANSAC is indeed feasible. Furthermore, we introduce a more strategic and computationally efficient approach to enhance the true detection rate and overall performance of the CTRL-RANSAC. Experiments conducted on synthetic and real-world datasets robustly support our theoretical results, showcasing the superior performance of the proposed method.
Authors: David Schulte, Felix Hamborg, Alan Akbik
Abstract: Intermediate task transfer learning can greatly improve model performance. If, for example, one has little training data for emotion detection, first fine-tuning a language model on a sentiment classification dataset may improve performance strongly. But which task to choose for transfer learning? Prior methods producing useful task rankings are infeasible for large source pools, as they require forward passes through all source language models. We overcome this by introducing Embedding Space Maps (ESMs), light-weight neural networks that approximate the effect of fine-tuning a language model. We conduct the largest study on NLP task transferability and task selection with 12k source-target pairs. We find that applying ESMs on a prior method reduces execution time and disk space usage by factors of 10 and 278, respectively, while retaining high selection performance (avg. regret@5 score of 2.95).
Authors: Srijan Saket, Mohit Agarwal, Rishabh Mehrotra
Abstract: The rise in popularity of social media platforms, has resulted in millions of new, content pieces being created every day. This surge in content creation underscores the need to pay attention to our design choices as they can greatly impact how long content remains relevant. In today's landscape where regularly recommending new content is crucial, particularly in the absence of detailed information, a variety of factors such as UI features, algorithms and system settings contribute to shaping the journey of content across the platform. While previous research has focused on how new content affects users' experiences, this study takes a different approach by analyzing these decisions considering the content itself. Through a series of carefully crafted experiments we explore how seemingly small decisions can influence the longevity of content, measured by metrics like Content Progression (CVP) and Content Survival (CSR). We also emphasize the importance of recognizing the stages that content goes through underscoring the need to tailor strategies for each stage as a one size fits all approach may not be effective. Additionally we argue for a departure from traditional experimental setups in the study of content lifecycles, to avoid potential misunderstandings while proposing advanced techniques, to achieve greater precision and accuracy in the evaluation process.
Authors: Gokul Puthumanaillam, Paulo Padrao, Jose Fuentes, Leonardo Bobadilla, Melkior Ornik
Abstract: Robots performing navigation tasks in complex environments face significant challenges due to uncertainty in state estimation. Effectively managing this uncertainty is crucial, but the optimal approach varies depending on the specific details of the task: different tasks require varying levels of precision in different regions of the environment. For instance, a robot navigating a crowded space might need precise localization near obstacles but can operate effectively with less precise state estimates in open areas. This varying need for certainty in different parts of the environment, depending on the task, calls for policies that can adapt their uncertainty management strategies based on task-specific requirements. In this paper, we present a framework for integrating task-specific uncertainty requirements directly into navigation policies. We introduce Task-Specific Uncertainty Map (TSUM), which represents acceptable levels of state estimation uncertainty across different regions of the operating environment for a given task. Using TSUM, we propose Generalized Uncertainty Integration for Decision-Making and Execution (GUIDE), a policy conditioning framework that incorporates these uncertainty requirements into the robot's decision-making process. We find that conditioning policies on TSUMs provides an effective way to express task-specific uncertainty requirements and enables the robot to reason about the context-dependent value of certainty. We show how integrating GUIDE into reinforcement learning frameworks allows the agent to learn navigation policies without the need for explicit reward engineering to balance task completion and uncertainty management. We evaluate GUIDE on a variety of real-world navigation tasks and find that it demonstrates significant improvements in task completion rates compared to baselines. Evaluation videos can be found at https://guided-agents.github.io.
Authors: Xin Liu, Weijia Zhang, Min-Ling Zhang
Abstract: In survival analysis, subjects often face competing risks; for example, individuals with cancer may also suffer from heart disease or other illnesses, which can jointly influence the prognosis of risks and censoring. Traditional survival analysis methods often treat competing risks as independent and fail to accommodate the dependencies between different conditions. In this paper, we introduce HACSurv, a survival analysis method that learns Hierarchical Archimedean Copulas structures and cause-specific survival functions from data with competing risks. HACSurv employs a flexible dependency structure using hierarchical Archimedean copulas to represent the relationships between competing risks and censoring. By capturing the dependencies between risks and censoring, HACSurv achieves better survival predictions and offers insights into risk interactions. Experiments on synthetic datasets demonstrate that our method can accurately identify the complex dependency structure and precisely predict survival distributions, whereas the compared methods exhibit significant deviations between their predictions and the true distributions. Experiments on multiple real-world datasets also demonstrate that our method achieves better survival prediction compared to previous state-of-the-art methods.
Authors: Baiyue Wang, Anthony Bloch
Abstract: We consider learning the dynamics of a typical nonholonomic system -- the rolling penny. A nonholonomic system is a system subject to nonholonomic constraints. Unlike holonomic constraints, a nonholonomic constraint does not define a submanifold on the configuration space. Therefore, the inverse problem of finding the constraints has to involve the tangent space. This paper discuss how to learn the dynamics, as well as the constraints for such a system given the data set of discrete trajectories on the tangent bundle $TQ$.
Authors: Wassim Jabbour, Enzo Benoit-Jeannin, Oscar Bedford, Saif Shahin
Abstract: Chess involves extensive study and requires players to keep manual records of their matches, a process which is time-consuming and distracting. The lack of high-quality labeled photographs of chess boards, and the tediousness of manual labeling, have hindered the wide application of Deep Learning (DL) to automating this record-keeping process. This paper proposes an end-to-end pipeline that employs domain adaptation (DA) to predict the labels of real, top-view, unlabeled chessboard images using synthetic, labeled images. The pipeline is composed of a pre-processing phase which detects the board, crops the individual squares, and feeds them one at a time to a DL model. The model then predicts the labels of the squares and passes the ordered predictions to a post-processing pipeline which generates the Forsyth-Edwards Notation (FEN) of the position. The three approaches considered are the following: A VGG16 model pre-trained on ImageNet, defined here as the Base-Source model, fine-tuned to predict source domain squares and then used to predict target domain squares without any domain adaptation; an improved version of the Base-Source model which applied CORAL loss to some of the final fully connected layers of the VGG16 to implement DA; and a Domain Adversarial Neural Network (DANN) which used the adversarial training of a domain discriminator to perform the DA. Also, although we opted not to use the labels of the target domain for this study, we trained a baseline with the same architecture as the Base-Source model (Named Base-Target) directly on the target domain in order to get an upper bound on the performance achievable through domain adaptation. The results show that the DANN model only results in a 3% loss in accuracy when compared to the Base-Target model while saving all the effort required to label the data.
Authors: Chandrajit Bajaj, Minh Nguyen, Shubham Bhardwaj
Abstract: Material segmentation is a complex task, particularly when dealing with aerial data in poor lighting and atmospheric conditions. To address this, hyperspectral data from specialized cameras can be very useful in addition to RGB images. However, due to hardware constraints, high spectral data often come with lower spatial resolution. Additionally, incorporating such data into a learning-based segmentation framework is challenging due to the numerous data channels involved. To overcome these difficulties, we propose an innovative Siamese framework that uses time series-based compression to effectively and scalably integrate the additional spectral data into the segmentation task. We demonstrate our model's effectiveness through competitive benchmarks on aerial datasets in various environmental conditions.
Authors: Subramanyam Sahoo, Kamlesh Dutta
Abstract: The relentless pursuit of technological advancements has ushered in a new era where artificial intelligence (AI) is not only a powerful tool but also a critical economic driver. At the forefront of this transformation is Generative AI, which is catalyzing a paradigm shift across industries. Deep generative models, an integration of generative and deep learning techniques, excel in creating new data beyond analyzing existing ones, revolutionizing sectors from production and manufacturing to finance. By automating design, optimization, and innovation cycles, Generative AI is reshaping core industrial processes. In the financial sector, it is transforming risk assessment, trading strategies, and forecasting, demonstrating its profound impact. This paper explores the sweeping changes driven by deep learning models like Large Language Models (LLMs), highlighting their potential to foster innovative business models, disruptive technologies, and novel economic landscapes. As we stand at the threshold of an AI-driven economic era, Generative AI is emerging as a pivotal force, driving innovation, disruption, and economic evolution on a global scale.
Authors: Zhen Qin, Zhihui Zhu
Abstract: Tensor train (TT) decomposition represents an $N$-order tensor using $O(N)$ matrices (i.e., factors) of small dimensions, achieved through products among these factors. Due to its compact representation, TT decomposition has found wide applications, including various tensor recovery problems in signal processing and quantum information. In this paper, we study the problem of reconstructing a TT format tensor from measurements that are contaminated by outliers with arbitrary values. Given the vulnerability of smooth formulations to corruptions, we use an $\ell_1$ loss function to enhance robustness against outliers. We first establish the $\ell_1/\ell_2$-restricted isometry property (RIP) for Gaussian measurement operators, demonstrating that the information in the TT format tensor can be preserved using a number of measurements that grows linearly with $N$. We also prove the sharpness property for the $\ell_1$ loss function optimized over TT format tensors. Building on the $\ell_1/\ell_2$-RIP and sharpness property, we then propose two complementary methods to recover the TT format tensor from the corrupted measurements: the projected subgradient method (PSubGM), which optimizes over the entire tensor, and the factorized Riemannian subgradient method (FRSubGM), which optimizes directly over the factors. Compared to PSubGM, the factorized approach FRSubGM significantly reduces the memory cost at the expense of a slightly slower convergence rate. Nevertheless, we show that both methods, with diminishing step sizes, converge linearly to the ground-truth tensor given an appropriate initialization, which can be obtained by a truncated spectral method.
Authors: Yuzhu Li, Hao Li, Weijie Chen, Keelan O'Riordan, Neha Mani, Yuxuan Qi, Tairan Liu, Sridhar Mani, Aydogan Ozcan
Abstract: Distinguishing between swarming and swimming, the two principal forms of bacterial movement, holds significant conceptual and clinical relevance. This is because bacteria that exhibit swarming capabilities often possess unique properties crucial to the pathogenesis of infectious diseases and may also have therapeutic potential. Here, we report a deep learning-based swarming classifier that rapidly and autonomously predicts swarming probability using a single blurry image. Compared with traditional video-based, manually-processed approaches, our method is particularly suited for high-throughput environments and provides objective, quantitative assessments of swarming probability. The swarming classifier demonstrated in our work was trained on Enterobacter sp. SM3 and showed good performance when blindly tested on new swarming (positive) and swimming (negative) test images of SM3, achieving a sensitivity of 97.44% and a specificity of 100%. Furthermore, this classifier demonstrated robust external generalization capabilities when applied to unseen bacterial species, such as Serratia marcescens DB10 and Citrobacter koseri H6. It blindly achieved a sensitivity of 97.92% and a specificity of 96.77% for DB10, and a sensitivity of 100% and a specificity of 97.22% for H6. This competitive performance indicates the potential to adapt our approach for diagnostic applications through portable devices or even smartphones. This adaptation would facilitate rapid, objective, on-site screening for bacterial swarming motility, potentially enhancing the early detection and treatment assessment of various diseases, including inflammatory bowel diseases (IBD) and urinary tract infections (UTI).
Authors: Benji Peng, Ziqian Bi, Qian Niu, Ming Liu, Pohsun Feng, Tianyang Wang, Lawrence K. Q. Yan, Yizhu Wen, Yichao Zhang, Caitlyn Heqi Yin
Abstract: Large Language Models (LLMs) have transformed artificial intelligence by advancing natural language understanding and generation, enabling applications across fields beyond healthcare, software engineering, and conversational systems. Despite these advancements in the past few years, LLMs have shown considerable vulnerabilities, particularly to prompt injection and jailbreaking attacks. This review analyzes the state of research on these vulnerabilities and presents available defense strategies. We roughly categorize attack approaches into prompt-based, model-based, multimodal, and multilingual, covering techniques such as adversarial prompting, backdoor injections, and cross-modality exploits. We also review various defense mechanisms, including prompt filtering, transformation, alignment techniques, multi-agent defenses, and self-regulation, evaluating their strengths and shortcomings. We also discuss key metrics and benchmarks used to assess LLM safety and robustness, noting challenges like the quantification of attack success in interactive contexts and biases in existing datasets. Identifying current research gaps, we suggest future directions for resilient alignment strategies, advanced defenses against evolving attacks, automation of jailbreak detection, and consideration of ethical and societal impacts. This review emphasizes the need for continued research and cooperation within the AI community to enhance LLM security and ensure their safe deployment.
Authors: Paul Goyes-Pe\~nafiel, Umair bin Waheed, Henry Arguello
Abstract: The global demand for unconventional energy sources such as geothermal energy and white hydrogen requires new exploration techniques for precise subsurface structure characterization and potential reservoir identification. Magnetotelluric (MT) inversion is crucial for these tasks, providing critical information on the distribution of subsurface electrical resistivity at depths ranging from hundreds to thousands of meters. However, traditional iterative algorithm-based inversion methods require the adjustment of multiple parameters, demanding time-consuming and exhaustive tuning processes to achieve proper cost function minimization. Although recent advances have incorporated deep learning algorithms for MT inversion, these have been primarily based on supervised learning, which needs large labeled datasets for training. Therefore, it causes issues in generalization and model characteristics that are restricted to the neural network's features. This work utilizes TensorFlow operations to create a differentiable forward MT operator, leveraging its automatic differentiation capability. Moreover, instead of solving for the subsurface model directly, as classical algorithms perform, this paper presents a new deep unsupervised inversion algorithm guided by physics to estimate 1D MT models. Instead of using datasets with the observed data and their respective model as labels during training, our method employs a differentiable modeling operator that physically guides the cost function minimization, making the proposed method solely dependent on observed data. Therefore, the optimization problem is updating the network weights to minimize the data misfit. We test the proposed method with field and synthetic data at different acquisition frequencies, demonstrating that the resistivity models are more accurate than other results using state-of-the-art techniques.
Authors: Yi Yang, Richard M. Voyles, Haiyan H. Zhang, Robert A. Nawrocki
Abstract: Accumulated detailed knowledge about the neuronal activities in human brains has brought more attention to bio-inspired spiking neural networks (SNNs). In contrast to non-spiking deep neural networks (DNNs), SNNs can encode and transmit spatiotemporal information more efficiently by exploiting biologically realistic and low-power event-driven neuromorphic architectures. However, the supervised learning of SNNs still remains a challenge because the spike-timing-dependent plasticity (STDP) of connected spiking neurons is difficult to implement and interpret in existing backpropagation learning schemes. This paper proposes a fractional-order spike-timing-dependent gradient descent (FO-STDGD) learning model by considering a derived nonlinear activation function that describes the relationship between the quasi-instantaneous firing rate and the temporal membrane potentials of nonleaky integrate-and-fire neurons. The training strategy can be generalized to any fractional orders between 0 and 2 since the FO-STDGD incorporates the fractional gradient descent method into the calculation of spike-timing-dependent loss gradients. The proposed FO-STDGD model is tested on the MNIST and DVS128 Gesture datasets and its accuracy under different network structure and fractional orders is analyzed. It can be found that the classification accuracy increases as the fractional order increases, and specifically, the case of fractional order 1.9 improves by 155% relative to the case of fractional order 1 (traditional gradient descent). In addition, our scheme demonstrates the state-of-the-art computational efficacy for the same SNN structure and training epochs.
Authors: Tianyi Xu, Patrick Zheng, Shiyan Liu, Sicheng Lyu, Isabeau Pr\'emont-Schwarz
Abstract: Artificial Neural Networks (ANNs) suffer from catastrophic forgetting, where the learning of new tasks causes the catastrophic forgetting of old tasks. Existing Machine Learning (ML) algorithms, including those using Stochastic Gradient Descent (SGD) and Hebbian Learning typically update their weights linearly with experience i.e., independently of their current strength. This contrasts with biological neurons, which at intermediate strengths are very plastic, but consolidate with Long-Term Potentiation (LTP) once they reach a certain strength. We hypothesize this mechanism might help mitigate catastrophic forgetting. We introduce Sigmoidal Neuronal Adaptive Plasticity (SNAP) an artificial approximation to Long-Term Potentiation for ANNs by having the weights follow a sigmoidal growth behaviour allowing the weights to consolidate and stabilize when they reach sufficiently large or small values. We then compare SNAP to linear weight growth and exponential weight growth and see that SNAP completely prevents the forgetting of previous tasks for Hebbian Learning but not for SGD-base learning.
Authors: Paul E. Chang, Nasrulloh Loka, Daolang Huang, Ulpu Remes, Samuel Kaski, Luigi Acerbi
Abstract: Amortized meta-learning methods based on pre-training have propelled fields like natural language processing and vision. Transformer-based neural processes and their variants are leading models for probabilistic meta-learning with a tractable objective. Often trained on synthetic data, these models implicitly capture essential latent information in the data-generation process. However, existing methods do not allow users to flexibly inject (condition on) and extract (predict) this probabilistic latent information at runtime, which is key to many tasks. We introduce the Amortized Conditioning Engine (ACE), a new transformer-based meta-learning model that explicitly represents latent variables of interest. ACE affords conditioning on both observed data and interpretable latent variables, the inclusion of priors at runtime, and outputs predictive distributions for discrete and continuous data and latents. We show ACE's modeling flexibility and performance in diverse tasks such as image completion and classification, Bayesian optimization, and simulation-based inference.
Authors: Zhekun Shi, Longlin Yu, Tianyu Xie, Cheng Zhang
Abstract: Recent success of diffusion models has inspired a surge of interest in developing sampling techniques using reverse diffusion processes. However, accurately estimating the drift term in the reverse stochastic differential equation (SDE) solely from the unnormalized target density poses significant challenges, hindering existing methods from achieving state-of-the-art performance. In this paper, we introduce the Diffusion-PINN Sampler (DPS), a novel diffusion-based sampling algorithm that estimates the drift term by solving the governing partial differential equation of the log-density of the underlying SDE marginals via physics-informed neural networks (PINN). We prove that the error of log-density approximation can be controlled by the PINN residual loss, enabling us to establish convergence guarantees of DPS. Experiments on a variety of sampling tasks demonstrate the effectiveness of our approach, particularly in accurately identifying mixing proportions when the target contains isolated components.
Authors: Yulin Song, Guorui Sang, Jing Yu, Chuangbai Xiao
Abstract: Singing voice synthesis (SVS) system is expected to generate high-fidelity singing voice from given music scores (lyrics, duration and pitch). Recently, diffusion models have performed well in this field. However, sacrificing inference speed to exchange with high-quality sample generation limits its application scenarios. In order to obtain high quality synthetic singing voice more efficiently, we propose a singing voice synthesis method based on the consistency model, ConSinger, to achieve high-fidelity singing voice synthesis with minimal steps. The model is trained by applying consistency constraint and the generation quality is greatly improved at the expense of a small amount of inference speed. Our experiments show that ConSinger is highly competitive with the baseline model in terms of generation speed and quality. Audio samples are available at https://keylxiao.github.io/consinger.
Authors: Yuki Kanto, Kohei Watabe
Abstract: In recent years, various services have been provided through high-speed and high-capacity wireless networks on mobile communication devices, necessitating stable communication regardless of indoor or outdoor environments. To achieve stable communication, it is essential to implement proactive measures, such as switching to an alternative path and ensuring data buffering before the communication quality becomes unstable. The technology of Wireless Link Quality Estimation (WLQE), which predicts the communication quality of wireless networks in advance, plays a crucial role in this context. In this paper, we propose a novel WLQE model for estimating the communication quality of wireless networks by leveraging sequential information. Our proposed method is based on Long Short-Term Memory (LSTM), enabling highly accurate estimation by considering the sequential information of link quality. We conducted a comparative evaluation with the conventional model, stacked autoencoder-based link quality estimator (LQE-SAE), using a dataset recorded in real-world environmental conditions. Our LSTM-based LQE model demonstrates its superiority, achieving a 4.0% higher accuracy and a 4.6% higher macro-F1 score than the LQE-SAE model in the evaluation.
Authors: Han Zhou, Jordy Van Landeghem, Teodora Popordanoska, Matthew B. Blaschko
Abstract: The selective classifier (SC) has garnered increasing interest in areas such as medical diagnostics, autonomous driving, and the justice system. The Area Under the Risk-Coverage Curve (AURC) has emerged as the foremost evaluation metric for assessing the performance of SC systems. In this work, we introduce a more straightforward representation of the population AURC, interpretable as a weighted risk function, and propose a Monte Carlo plug-in estimator applicable to finite sample scenarios. We demonstrate that our estimator is consistent and offers a low-bias estimation of the actual weights, with a tightly bounded mean squared error (MSE). We empirically show the effectiveness of this estimator on a comprehensive benchmark across multiple datasets, model architectures, and Confidence Score Functions (CSFs).
Authors: Wojciech Anyszka, Kaja Gruntkowska, Alexander Tyurin, Peter Richt\'arik
Abstract: We revisit FedExProx - a recently proposed distributed optimization method designed to enhance convergence properties of parallel proximal algorithms via extrapolation. In the process, we uncover a surprising flaw: its known theoretical guarantees on quadratic optimization tasks are no better than those offered by the vanilla Gradient Descent (GD) method. Motivated by this observation, we develop a novel analysis framework, establishing a tighter linear convergence rate for non-strongly convex quadratic problems. By incorporating both computation and communication costs, we demonstrate that FedExProx can indeed provably outperform GD, in stark contrast to the original analysis. Furthermore, we consider partial participation scenarios and analyze two adaptive extrapolation strategies - based on gradient diversity and Polyak stepsizes - again significantly outperforming previous results. Moving beyond quadratics, we extend the applicability of our analysis to general functions satisfying the Polyak-Lojasiewicz condition, outperforming the previous strongly convex analysis while operating under weaker assumptions. Backed by empirical results, our findings point to a new and stronger potential of FedExProx, paving the way for further exploration of the benefits of extrapolation in federated learning.
Authors: Yuji Wang, Zehua Chen, Xiaoyu Chen, Jun Zhu, Jianfei Chen
Abstract: Image-to-video (I2V) generation is gaining increasing attention with its wide application in video synthesis. Recently, diffusion-based I2V models have achieved remarkable progress given their novel design on network architecture, cascaded framework, and motion representation. However, restricted by their noise-to-data generation process, diffusion-based methods inevitably suffer the difficulty to generate video samples with both appearance consistency and temporal coherence from an uninformative Gaussian noise, which may limit their synthesis quality. In this work, we present FrameBridge, taking the given static image as the prior of video target and establishing a tractable bridge model between them. By formulating I2V synthesis as a frames-to-frames generation task and modelling it with a data-to-data process, we fully exploit the information in input image and facilitate the generative model to learn the image animation process. In two popular settings of training I2V models, namely fine-tuning a pre-trained text-to-video (T2V) model or training from scratch, we further propose two techniques, SNR-Aligned Fine-tuning (SAF) and neural prior, which improve the fine-tuning efficiency of diffusion-based T2V models to FrameBridge and the synthesis quality of bridge-based I2V models respectively. Experiments conducted on WebVid-2M and UCF-101 demonstrate that: (1) our FrameBridge achieves superior I2V quality in comparison with the diffusion counterpart (zero-shot FVD 83 vs. 176 on MSR-VTT and non-zero-shot FVD 122 vs. 171 on UCF-101); (2) our proposed SAF and neural prior effectively enhance the ability of bridge-based I2V models in the scenarios of fine-tuning and training from scratch. Demo samples can be visited at: https://framebridge-demo.github.io/.
Authors: Abdur Rashid, Parag Biswas, abdullah al masum, MD Abdullah Al Nasim, Kishor Datta Gupta
Abstract: The integration of machine learning into smart grid systems represents a transformative step in enhancing the efficiency, reliability, and sustainability of modern energy networks. By adding advanced data analytics, these systems can better manage the complexities of renewable energy integration, demand response, and predictive maintenance. Machine learning algorithms analyze vast amounts of data from smart meters, sensors, and other grid components to optimize energy distribution, forecast demand, and detect irregularities that could indicate potential failures. This enables more precise load balancing, reduces operational costs, and enhances the resilience of the grid against disturbances. Furthermore, the use of predictive models helps in anticipating equipment failures, thereby improving the reliability of the energy supply. As smart grids continue to evolve, the role of machine learning in managing decentralized energy sources and enabling real-time decision-making will become increasingly critical. However, the deployment of these technologies also raises challenges related to data privacy, security, and the need for robust infrastructure. Addressing these issues in this research authors will focus on realizing the full potential of smart grids, ensuring they meet the growing energy demands while maintaining a focus on sustainability and efficiency using Machine Learning techniques. Furthermore, this research will help determine the smart grid's essentiality with the aid of Machine Learning. Multiple ML algorithms have been integrated along with their pros and cons. The future scope of these algorithms are also integrated.
Authors: Mogens Plessen
Abstract: This paper addresses two tasks: (i) fixed-size objects such as hay bales are to be identified in an aerial image for a given reference image of the object, and (ii) variable-size patches such as areas on fields requiring spot spraying or other handling are to be identified in an image for a given small-scale reference image. Both tasks are related. The second differs in that identified sub-images similar to the reference image are further clustered before patches contours are determined by solving a traveling salesman problem. Both tasks are complex in that the exact number of similar sub-images is not known a priori. The main discussion of this paper is presentation of an acceleration mechanism for sub-image search that is based on a transformation of an image to multivariate time series along the RGB-channels and subsequent segmentation to reduce the 2D search space in the image. Two variations of the acceleration mechanism are compared to exhaustive search on diverse synthetic and real-world images. Quantitatively, proposed method results in solve time reductions of up to 2 orders of magnitude, while qualitatively delivering comparative visual results. Proposed method is neural network-free and does not use any image pre-processing.
Authors: Maor Biton Dor, Yisroel Mirsky
Abstract: This paper introduces a novel data-free model extraction attack that significantly advances the current state-of-the-art in terms of efficiency, accuracy, and effectiveness. Traditional black-box methods rely on using the victim's model as an oracle to label a vast number of samples within high-confidence areas. This approach not only requires an extensive number of queries but also results in a less accurate and less transferable model. In contrast, our method innovates by focusing on sampling low-confidence areas (along the decision boundaries) and employing an evolutionary algorithm to optimize the sampling process. These novel contributions allow for a dramatic reduction in the number of queries needed by the attacker by a factor of 10x to 600x while simultaneously improving the accuracy of the stolen model. Moreover, our approach improves boundary alignment, resulting in better transferability of adversarial examples from the stolen model to the victim's model (increasing the attack success rate from 60\% to 82\% on average). Finally, we accomplish all of this with a strict black-box assumption on the victim, with no knowledge of the target's architecture or dataset. We demonstrate our attack on three datasets with increasingly larger resolutions and compare our performance to four state-of-the-art model extraction attacks.
Authors: Jenelle Feather, David Lipshutz, Sarah E. Harvey, Alex H. Williams, Eero P. Simoncelli
Abstract: Image representations (artificial or biological) are often compared in terms of their global geometry; however, representations with similar global structure can have strikingly different local geometries. Here, we propose a framework for comparing a set of image representations in terms of their local geometries. We quantify the local geometry of a representation using the Fisher information matrix, a standard statistical tool for characterizing the sensitivity to local stimulus distortions, and use this as a substrate for a metric on the local geometry in the vicinity of a base image. This metric may then be used to optimally differentiate a set of models, by finding a pair of "principal distortions" that maximize the variance of the models under this metric. We use this framework to compare a set of simple models of the early visual system, identifying a novel set of image distortions that allow immediate comparison of the models by visual inspection. In a second example, we apply our method to a set of deep neural network models and reveal differences in the local geometry that arise due to architecture and training types. These examples highlight how our framework can be used to probe for informative differences in local sensitivities between complex computational models, and suggest how it could be used to compare model representations with human perception.
Authors: Sarah Tan, Keri Mallari, Julius Adebayo, Albert Gordo, Martin T. Wells, Kori Inkpen
Abstract: Large generative models (LMs) are increasingly being considered for high-stakes decision-making. This work considers how such models compare to humans and predictive AI models on a specific case of recidivism prediction. We combine three datasets -- COMPAS predictive AI risk scores, human recidivism judgements, and photos -- into a dataset on which we study the properties of several state-of-the-art, multimodal LMs. Beyond accuracy and bias, we focus on studying human-LM alignment on the task of recidivism prediction. We investigate if these models can be steered towards human decisions, the impact of adding photos, and whether anti-discimination prompting is effective. We find that LMs can be steered to outperform humans and COMPAS using in context-learning. We find anti-discrimination prompting to have unintended effects, causing some models to inhibit themselves and significantly reduce their number of positive predictions.
Authors: Fnu Neha, Arvind K. Bansal
Abstract: Renal tumors, especially renal cell carcinoma (RCC), show significant heterogeneity, posing challenges for diagnosis using radiology images such as MRI, echocardiograms, and CT scans. U-Net based deep learning techniques are emerging as a promising approach for automated medical image segmentation for minimally invasive diagnosis of renal tumors. However, current techniques need further improvements in accuracy to become clinically useful to radiologists. In this study, we present an improved U-Net based model for end-to-end automated semantic segmentation of CT scan images to identify renal tumors. The model uses residual connections across convolution layers, integrates a multi-layer feature fusion (MFF) and cross-channel attention (CCA) within encoder blocks, and incorporates skip connections augmented with additional information derived using MFF and CCA. We evaluated our model on the KiTS19 dataset, which contains data from 210 patients. For kidney segmentation, our model achieves a Dice Similarity Coefficient (DSC) of 0.97 and a Jaccard index (JI) of 0.95. For renal tumor segmentation, our model achieves a DSC of 0.96 and a JI of 0.91. Based on a comparison of available DSC scores, our model outperforms the current leading models.
Authors: Jason Jabbour, Vijay Janapa Reddi
Abstract: The integration of Generative Artificial Intelligence (AI) into autonomous machines represents a major paradigm shift in how these systems operate and unlocks new solutions to problems once deemed intractable. Although generative AI agents provide unparalleled capabilities, they also have unique safety concerns. These challenges require robust safeguards, especially for autonomous machines that operate in high-stakes environments. This work investigates the evolving safety requirements when generative models are integrated as agents into physical autonomous machines, comparing these to safety considerations in less critical AI applications. We explore the challenges and opportunities to ensure the safe deployment of generative AI-driven autonomous machines. Furthermore, we provide a forward-looking perspective on the future of AI-driven autonomous systems and emphasize the importance of evaluating and communicating safety risks. As an important step towards addressing these concerns, we recommend the development and implementation of comprehensive safety scorecards for the use of generative AI technologies in autonomous machines.
Authors: Jerry Huang, Laura Lewis, Hsin-Yuan Huang, John Preskill
Abstract: Recent advances have demonstrated that $\mathcal{O}(\log M)$ measurements suffice to predict $M$ properties of arbitrarily large quantum many-body systems. However, these remarkable findings assume that the properties to be predicted are chosen independently of the data. This assumption can be violated in practice, where scientists adaptively select properties after looking at previous predictions. This work investigates the adaptive setting for three classes of observables: local, Pauli, and bounded-Frobenius-norm observables. We prove that $\Omega(\sqrt{M})$ samples of an arbitrarily large unknown quantum state are necessary to predict expectation values of $M$ adaptively chosen local and Pauli observables. We also present computationally-efficient algorithms that achieve this information-theoretic lower bound. In contrast, for bounded-Frobenius-norm observables, we devise an algorithm requiring only $\mathcal{O}(\log M)$ samples, independent of system size. Our results highlight the potential pitfalls of adaptivity in analyzing data from quantum experiments and provide new algorithmic tools to safeguard against erroneous predictions in quantum experiments.
Authors: Srishti Gureja, Lester James V. Miranda, Shayekh Bin Islam, Rishabh Maheshwary, Drishti Sharma, Gusti Winata, Nathan Lambert, Sebastian Ruder, Sara Hooker, Marzieh Fadaee
Abstract: Reward models (RMs) have driven the state-of-the-art performance of LLMs today by enabling the integration of human feedback into the language modeling process. However, RMs are primarily trained and evaluated in English, and their capabilities in multilingual settings remain largely understudied. In this work, we conduct a systematic evaluation of several reward models in multilingual settings. We first construct the first-of-its-kind multilingual RM evaluation benchmark, M-RewardBench, consisting of 2.87k preference instances for 23 typologically diverse languages, that tests the chat, safety, reasoning, and translation capabilities of RMs. We then rigorously evaluate a wide range of reward models on M-RewardBench, offering fresh insights into their performance across diverse languages. We identify a significant gap in RMs' performances between English and non-English languages and show that RM preferences can change substantially from one language to another. We also present several findings on how different multilingual aspects impact RM performance. Specifically, we show that the performance of RMs is improved with improved translation quality. Similarly, we demonstrate that the models exhibit better performance for high-resource languages. We release M-RewardBench dataset and the codebase in this study to facilitate a better understanding of RM evaluation in multilingual settings.
Authors: Mamadou K. Keita, Christopher Homan, Sofiane Abdoulaye Hamani, Adwoa Bremang, Marcos Zampieri, Habibatou Abdoulaye Alfari, Elysabhete Amadou Ibrahim, Dennis Owusu
Abstract: Grammatical error correction (GEC) is important for improving written materials for low-resource languages like Zarma -- spoken by over 5 million people in West Africa. Yet it remains a challenging problem. This study compares rule-based methods, machine translation (MT) models, and large language models (LLMs) for GEC in Zarma. We evaluate each approach's effectiveness on our manually-built dataset of over 250,000 examples using synthetic and human-annotated data. Our experiments show that the MT-based approach using the M2M100 model outperforms others, achieving a detection rate of 95.82% and a suggestion accuracy of 78.90% in automatic evaluations, and scoring 3.0 out of 5.0 in logical/grammar error correction during MEs by native speakers. The rule-based method achieved perfect detection (100%) and high suggestion accuracy (96.27%) for spelling corrections but struggled with context-level errors. LLMs like MT5-small showed moderate performance with a detection rate of 90.62% and a suggestion accuracy of 57.15%. Our work highlights the potential of MT models to enhance GEC in low-resource languages, paving the way for more inclusive NLP tools.
Authors: Amin Sarihi, Ahmad Patooghy, Peter Jamieson, Abdel-Hameed A. Badawy
Abstract: This work focuses on advancing security research in the hardware design space by formally defining the realistic problem of Hardware Trojan (HT) detection. The goal is to model HT detection more closely to the real world, i.e., describing the problem as The Seeker's Dilemma where a detecting agent is unaware of whether circuits are infected by HTs or not. Using this theoretical problem formulation, we create a benchmark that consists of a mixture of HT-free and HT-infected restructured circuits while preserving their original functionalities. The restructured circuits are randomly infected by HTs, causing a situation where the defender is uncertain if a circuit is infected or not. We believe that our innovative benchmark and methodology of creating benchmarks will help the community judge the detection quality of different methods by comparing their success rates in circuit classification. We use our developed benchmark to evaluate three state-of-the-art HT detection tools to show baseline results for this approach. We use Principal Component Analysis to assess the strength of our benchmark, where we observe that some restructured HT-infected circuits are mapped closely to HT-free circuits, leading to significant label misclassification by detectors.
Authors: Laurence Liang
Abstract: Recent advances show that large language models (LLMs) generalize strong performance across different natural language benchmarks. However, the large size of LLMs makes training and inference expensive and impractical to run in resource-limited settings. This paper introduces a new approach called fine-tuning stacks of language models (FSLM), which involves stacking small language models (SLM) as an alternative to LLMs. By fine-tuning each SLM to perform a specific task, this approach breaks down high level reasoning into multiple lower-level steps that specific SLMs are responsible for. As a result, FSLM allows for lower training and inference costs, and also improves model interpretability as each SLM communicates with the subsequent one through natural language. By evaluating FSLM on common natural language benchmarks, this paper highlights promising early results toward generalizable performance using FSLM as a cost-effective alternative to LLMs.
Authors: Junsik Kim, Zhiyi Shi, Davin Jeong, Johannes Knittel, Helen Y. Yang, Yonghyun Song, Wanhua Li, Yicong Li, Dalit Ben-Yosef, Daniel Needleman, Hanspeter Pfister
Abstract: In clinical In-Vitro Fertilization (IVF), identifying the most viable embryo for transfer is important to increasing the likelihood of a successful pregnancy. Traditionally, this process involves embryologists manually assessing embryos' static morphological features at specific intervals using light microscopy. This manual evaluation is not only time-intensive and costly, due to the need for expert analysis, but also inherently subjective, leading to variability in the selection process. To address these challenges, we develop a multimodal model that leverages both time-lapse video data and Electronic Health Records (EHRs) to predict embryo viability. One of the primary challenges of our research is to effectively combine time-lapse video and EHR data, owing to their inherent differences in modality. We comprehensively analyze our multimodal model with various modality inputs and integration approaches. Our approach will enable fast and automated embryo viability predictions in scale for clinical IVF.
Authors: Wenrui Gou, Wenhui Ge, YangTan, Guisheng Fan, Mingchen Li, Huiqun Yu
Abstract: Protein structures are important for understanding their functions and interactions. Currently, many protein structure prediction methods are enriching the structure database. Discriminating the origin of structures is crucial for distinguishing between experimentally resolved and computationally predicted structures, evaluating the reliability of prediction methods, and guiding downstream biological studies. Building on works in structure prediction, We developed a structure-sensitive supervised deep learning model, Crystal vs Predicted Evaluator for Protein Structure (CPE-Pro), to represent and discriminate the origin of protein structures. CPE-Pro learns the structural information of proteins and captures inter-structural differences to achieve accurate traceability on four data classes, and is expected to be extended to more. Simultaneously, we utilized Foldseek to encode protein structures into "structure-sequence" and trained a protein Structural Sequence Language Model, SSLM. Preliminary experiments demonstrated that, compared to large-scale protein language models pre-trained on vast amounts of amino acid sequences, the "structure-sequences" enable the language model to learn more informative protein features, enhancing and optimizing structural representations. We have provided the code, model weights, and all related materials on https://github.com/GouWenrui/CPE-Pro-main.git.
Authors: Wenyi Xiao, Zechuan Wang, Leilei Gan, Shuai Zhao, Wanggui He, Luu Anh Tuan, Long Chen, Hao Jiang, Zhou Zhao, Fei Wu
Abstract: With the rapid advancement of large language models (LLMs), aligning policy models with human preferences has become increasingly critical. Direct Preference Optimization (DPO) has emerged as a promising approach for alignment, acting as an RL-free alternative to Reinforcement Learning from Human Feedback (RLHF). Despite DPO's various advancements and inherent limitations, an in-depth review of these aspects is currently lacking in the literature. In this work, we present a comprehensive review of the challenges and opportunities in DPO, covering theoretical analyses, variants, relevant preference datasets, and applications. Specifically, we categorize recent studies on DPO based on key research questions to provide a thorough understanding of DPO's current landscape. Additionally, we propose several future research directions to offer insights on model alignment for the research community.
Authors: Ismail Bibers, Osvaldo Arreche, Mustafa Abdallah
Abstract: The escalating frequency of intrusions in networked systems has spurred the exploration of new research avenues in devising artificial intelligence (AI) techniques for intrusion detection systems (IDS). Various AI techniques have been used to automate network intrusion detection tasks, yet each model possesses distinct strengths and weaknesses. Selecting the optimal model for a given dataset can pose a challenge, necessitating the exploration of ensemble methods to enhance generalization and applicability in network intrusion detection. This paper addresses this gap by conducting a comprehensive evaluation of diverse individual models and both simple and advanced ensemble methods for network IDS. We introduce an ensemble learning framework tailored for assessing individual models and ensemble methods in network intrusion detection tasks. Our framework encompasses the loading of input datasets, training of individual models and ensemble methods, and the generation of evaluation metrics. Furthermore, we incorporate all features across individual models and ensemble techniques. The study presents results for our framework, encompassing 14 methods, including various bagging, stacking, blending, and boosting techniques applied to multiple base learners such as decision trees, neural networks, and among others. We evaluate the framework using two distinct network intrusion datasets, RoEduNet-SIMARGL2021 and CICIDS-2017, each possessing unique characteristics. Additionally, we categorize AI models based on their performances on our evaluation metrics and via their confusion matrices. Our assessment demonstrates the efficacy of learning across most setups explored in this study. Furthermore, we contribute to the community by releasing our source codes, providing a foundational ensemble learning framework for network intrusion detection.
Authors: Nat Jeffries, Evan King, Manjunath Kudlur, Guy Nicholson, James Wang, Pete Warden
Abstract: This paper introduces Moonshine, a family of speech recognition models optimized for live transcription and voice command processing. Moonshine is based on an encoder-decoder transformer architecture and employs Rotary Position Embedding (RoPE) instead of traditional absolute position embeddings. The model is trained on speech segments of various lengths, but without using zero-padding, leading to greater efficiency for the encoder during inference time. When benchmarked against OpenAI's Whisper tiny.en, Moonshine Tiny demonstrates a 5x reduction in compute requirements for transcribing a 10-second speech segment while incurring no increase in word error rates across standard evaluation datasets. These results highlight Moonshine's potential for real-time and resource-constrained applications.
Authors: Guanhang Lei, Zhen Lei, Lei Shi
Abstract: Neural operators have shown promise in solving many types of Partial Differential Equations (PDEs). They are significantly faster compared to traditional numerical solvers once they have been trained with a certain amount of observed data. However, their numerical performance in solving time-dependent PDEs, particularly in long-time prediction of dynamic systems, still needs improvement. In this paper, we focus on solving the long-time integration of nonlinear wave equations via neural operators by replacing the initial condition with the prediction in a recurrent manner. Given limited observed temporal trajectory data, we utilize some intrinsic features of these nonlinear wave equations, such as conservation laws and well-posedness, to improve the algorithm design and reduce accumulated error. Our numerical experiments examine these improvements in the Korteweg-de Vries (KdV) equation, the sine-Gordon equation, and a semilinear wave equation on the irregular domain.
Authors: Subhankar Ghosh, Arun Sharma, Jayant Gupta, Aneesh Subramanian, Shashi Shekhar
Abstract: Given coarser-resolution projections from global climate models or satellite data, the downscaling problem aims to estimate finer-resolution regional climate data, capturing fine-scale spatial patterns and variability. Downscaling is any method to derive high-resolution data from low-resolution variables, often to provide more detailed and local predictions and analyses. This problem is societally crucial for effective adaptation, mitigation, and resilience against significant risks from climate change. The challenge arises from spatial heterogeneity and the need to recover finer-scale features while ensuring model generalization. Most downscaling methods \cite{Li2020} fail to capture the spatial dependencies at finer scales and underperform on real-world climate datasets, such as sea-level rise. We propose a novel Kriging-informed Conditional Diffusion Probabilistic Model (Ki-CDPM) to capture spatial variability while preserving fine-scale features. Experimental results on climate data show that our proposed method is more accurate than state-of-the-art downscaling techniques.
Authors: Clara Na, Ian Magnusson, Ananya Harsh Jha, Tom Sherborne, Emma Strubell, Jesse Dodge, Pradeep Dasigi
Abstract: Training data compositions for Large Language Models (LLMs) can significantly affect their downstream performance. However, a thorough data ablation study exploring large sets of candidate data mixtures is typically prohibitively expensive since the full effect is seen only after training the models; this can lead practitioners to settle for sub-optimal data mixtures. We propose an efficient method for approximating data ablations which trains individual models on subsets of a training corpus and reuses them across evaluations of combinations of subsets. In continued pre-training experiments, we find that, given an arbitrary evaluation set, the perplexity score of a single model trained on a candidate set of data is strongly correlated with perplexity scores of parameter averages of models trained on distinct partitions of that data. From this finding, we posit that researchers and practitioners can conduct inexpensive simulations of data ablations by maintaining a pool of models that were each trained on partitions of a large training corpus, and assessing candidate data mixtures by evaluating parameter averages of combinations of these models. This approach allows for substantial improvements in amortized training efficiency -- scaling only linearly with respect to new data -- by enabling reuse of previous training computation, opening new avenues for improving model performance through rigorous, incremental data assessment and mixing.
Authors: Xun Jiang, Feng Li, Han Zhao, Jiaying Wang, Jun Shao, Shihao Xu, Shu Zhang, Weiling Chen, Xavier Tang, Yize Chen, Mengyue Wu, Weizhi Ma, Mengdi Wang, Tianqiao Chen
Abstract: Large language models (LLMs) like GPTs, trained on vast datasets, have demonstrated impressive capabilities in language understanding, reasoning, and planning, achieving human-level performance in various tasks. Most studies focus on enhancing these models by training on ever-larger datasets to build more powerful foundation models. While training stronger models is important, enabling models to evolve during inference is equally crucial, a process we refer to as AI self-evolution. Unlike large-scale training, self-evolution may rely on limited data or interactions. Inspired by the columnar organization of the human cerebral cortex, we hypothesize that AI models could develop cognitive abilities and build internal representations through iterative interactions with their environment. To achieve this, models need long-term memory (LTM) to store and manage processed interaction data. LTM supports self-evolution by representing diverse experiences across environments and agents. In this report, we explore AI self-evolution and its potential to enhance models during inference. We examine LTM's role in lifelong learning, allowing models to evolve based on accumulated interactions. We outline the structure of LTM and the systems needed for effective data retention and representation. We also classify approaches for building personalized models with LTM data and show how these models achieve self-evolution through interaction. Using LTM, our multi-agent framework OMNE achieved first place on the GAIA benchmark, demonstrating LTM's potential for AI self-evolution. Finally, we present a roadmap for future research, emphasizing the importance of LTM for advancing AI technology and its practical applications.
Authors: Changmao Li, Jeffrey Flanigan
Abstract: Large Language Models (LLMs) exhibit impressive results across a wide range of natural language processing (NLP) tasks, yet they can often produce factually incorrect outputs. This paper introduces a simple but effective low-latency post-correction method, \textbf{Retrieval Augmented Correction (RAC)}, aimed at enhancing the factual performance of LLMs without requiring additional fine-tuning. Our method is general and can be used with any instruction-tuned LLM, and has greatly reduced latency compared to prior approaches. RAC decomposes the LLM's output into atomic facts and applies a fine-grained verification and correction process with retrieved content to verify and correct the LLM-generated output. Our extensive experiments show that RAC yields up to 30\% improvements over state-of-the-art baselines across two popular factuality evaluation datasets, validating its efficacy and robustness in both with and without the integration of Retrieval-Augmented Generation (RAG) across different LLMs.\footnote{Our code is at \url{https://github.com/jlab-nlp/Retrieval-Augmented-Correction}}
URLs: https://github.com/jlab-nlp/Retrieval-Augmented-Correction
Authors: Shibo Zhou, Bo Yang, Mengwen Yuan, Runhao Jiang, Rui Yan, Gang Pan, Huajin Tang
Abstract: Spiking Neural Networks (SNNs), renowned for their low power consumption, brain-inspired architecture, and spatio-temporal representation capabilities, have garnered considerable attention in recent years. Similar to Artificial Neural Networks (ANNs), high-quality benchmark datasets are of great importance to the advances of SNNs. However, our analysis indicates that many prevalent neuromorphic datasets lack strong temporal correlation, preventing SNNs from fully exploiting their spatio-temporal representation capabilities. Meanwhile, the integration of event and frame modalities offers more comprehensive visual spatio-temporal information. Yet, the SNN-based cross-modality fusion remains underexplored. In this work, we present a neuromorphic dataset called DVS-SLR that can better exploit the inherent spatio-temporal properties of SNNs. Compared to existing datasets, it offers advantages in terms of higher temporal correlation, larger scale, and more varied scenarios. In addition, our neuromorphic dataset contains corresponding frame data, which can be used for developing SNN-based fusion methods. By virtue of the dual-modal feature of the dataset, we propose a Cross-Modality Attention (CMA) based fusion method. The CMA model efficiently utilizes the unique advantages of each modality, allowing for SNNs to learn both temporal and spatial attention scores from the spatio-temporal features of event and frame modalities, subsequently allocating these scores across modalities to enhance their synergy. Experimental results demonstrate that our method not only improves recognition accuracy but also ensures robustness across diverse scenarios.
Authors: Rapha\"el Carpintero Perez (CMAP), S\'ebastien da Veiga (ENSAI, CREST), Josselin Garnier (CMAP), Brian Staber
Abstract: In computational physics, machine learning has now emerged as a powerful complementary tool to explore efficiently candidate designs in engineering studies. Outputs in such supervised problems are signals defined on meshes, and a natural question is the extension of general scalar output regression models to such complex outputs. Changes between input geometries in terms of both size and adjacency structure in particular make this transition non-trivial. In this work, we propose an innovative strategy for Gaussian process regression where inputs are large and sparse graphs with continuous node attributes and outputs are signals defined on the nodes of the associated inputs. The methodology relies on the combination of regularized optimal transport, dimension reduction techniques, and the use of Gaussian processes indexed by graphs. In addition to enabling signal prediction, the main point of our proposal is to come with confidence intervals on node values, which is crucial for uncertainty quantification and active learning. Numerical experiments highlight the efficiency of the method to solve real problems in fluid dynamics and solid mechanics.
Authors: Cristian Meo, Akihiro Nakano, Mircea Lic\u{a}, Aniket Didolkar, Masahiro Suzuki, Anirudh Goyal, Mengmi Zhang, Justin Dauwels, Yutaka Matsuo, Yoshua Bengio
Abstract: Unsupervised object-centric learning from videos is a promising approach towards learning compositional representations that can be applied to various downstream tasks, such as prediction and reasoning. Recently, it was shown that pretrained Vision Transformers (ViTs) can be useful to learn object-centric representations on real-world video datasets. However, while these approaches succeed at extracting objects from the scenes, the slot-based representations fail to maintain temporal consistency across consecutive frames in a video, i.e. the mapping of objects to slots changes across the video. To address this, we introduce Conditional Autoregressive Slot Attention (CA-SA), a framework that enhances the temporal consistency of extracted object-centric representations in video-centric vision tasks. Leveraging an autoregressive prior network to condition representations on previous timesteps and a novel consistency loss function, CA-SA predicts future slot representations and imposes consistency across frames. We present qualitative and quantitative results showing that our proposed method outperforms the considered baselines on downstream tasks, such as video prediction and visual question-answering tasks.
Authors: Montreuil Yannis, Yeo Shu Heng, Carlier Axel, Ng Lai Xing, Ooi Wei Tsang
Abstract: The Learning-to-Defer approach has been explored for classification and, more recently, regression tasks separately. Many contemporary learning tasks, however, involves both classification and regression components. In this paper, we introduce a Learning-to-Defer approach for multi-task learning that encompasses both classification and regression tasks. Our two-stage approach utilizes a rejector that defers decisions to the most accurate agent among a pre-trained joint classifier-regressor models and one or more external experts. We show that our surrogate loss is $(\mathcal{H}, \mathcal{F}, \mathcal{R})$ and Bayes--consistent, ensuring an effective approximation of the optimal solution. Additionally, we derive learning bounds that demonstrate the benefits of employing multiple confident experts along a rich model in a two-stage learning framework. Empirical experiments conducted on electronic health record analysis tasks underscore the performance enhancements achieved through our method.
Authors: Montreuil Yannis, Carlier Axel, Ng Lai Xing, Ooi Wei Tsang
Abstract: Pre-trained language models have profoundly impacted the field of extractive question-answering, leveraging large-scale textual corpora to enhance contextual language understanding. Despite their success, these models struggle in complex scenarios that demand nuanced interpretation or inferential reasoning beyond immediate textual cues. Furthermore, their size poses deployment challenges on resource-constrained devices. Addressing these limitations, we introduce an adapted two-stage Learning-to-Defer mechanism that enhances decision-making by enabling selective deference to human experts or larger models without retraining language models in the context of question-answering. This approach not only maintains computational efficiency but also significantly improves model reliability and accuracy in ambiguous contexts. We establish the theoretical soundness of our methodology by proving Bayes and $(\mathcal{H}, \mathcal{R})$--consistency of our surrogate loss function, guaranteeing the optimality of the final solution. Empirical evaluations on the SQuADv2 dataset illustrate performance gains from integrating human expertise and leveraging larger models. Our results further demonstrate that deferring a minimal number of queries allows the smaller model to achieve performance comparable to their larger counterparts while preserving computing efficiency, thus broadening the applicability of pre-trained language models in diverse operational environments.
Authors: Tianlin Liu, Jannes M\"unchmeyer, Laura Laurenti, Chris Marone, Maarten V. de Hoop, Ivan Dokmani\'c
Abstract: We introduce the Seismic Language Model (SeisLM), a foundational model designed to analyze seismic waveforms -- signals generated by Earth's vibrations such as the ones originating from earthquakes. SeisLM is pretrained on a large collection of open-source seismic datasets using a self-supervised contrastive loss, akin to BERT in language modeling. This approach allows the model to learn general seismic waveform patterns from unlabeled data without being tied to specific downstream tasks. When fine-tuned, SeisLM excels in seismological tasks like event detection, phase-picking, onset time regression, and foreshock-aftershock classification. The code has been made publicly available on https://github.com/liutianlin0121/seisLM.
Authors: Sheng Liu, Haotian Ye, James Zou
Abstract: Hallucination poses a challenge to the deployment of large vision-language models (LVLMs) in applications. Unlike in large language models (LLMs), hallucination in LVLMs often arises from misalignments between visual inputs and textual outputs. This paper investigates the underlying mechanisms of hallucination, focusing on the unique structure of LVLMs that distinguishes them from large language models (LLMs). We identify that hallucinations often arise from the sensitivity of text decoders to vision inputs, a natural phenomenon when image encoders and text decoders are pre-trained separately. Inspired by this, we introduce Visual and Textual Intervention (VTI), a novel technique designed to reduce hallucinations by steering latent space representations during inference to enhance the stability of vision features. As a task-agnostic test-time intervention, VTI can be easily applied to any problem without additional cost. Extensive experiments demonstrate that it can effectively reduce hallucinations and outperform baseline methods across multiple metrics, highlighting the critical role of vision feature stability in LVLMs.
Authors: Edi Sutoyo, Paris Avgeriou, Andrea Capiluppi
Abstract: Self-Admitted Technical Debt (SATD) refers to circumstances where developers use textual artifacts to explain why the existing implementation is not optimal. Past research in detecting SATD has focused on either identifying SATD (classifying SATD items as SATD or not) or categorizing SATD (labeling instances as SATD that pertain to requirement, design, code, test debt, etc.). However, the performance of these approaches remains suboptimal, particularly for specific types of SATD, such as test and requirement debt, primarily due to extremely imbalanced datasets. To address these challenges, we build on earlier research by utilizing BiLSTM architecture for the binary identification of SATD and BERT architecture for categorizing different types of SATD. Despite their effectiveness, both architectures struggle with imbalanced data. Therefore, we employ a large language model data augmentation strategy to mitigate this issue. Furthermore, we introduce a two-step approach to identify and categorize SATD across various datasets derived from different artifacts. Our contributions include providing a balanced dataset for future SATD researchers and demonstrating that our approach significantly improves SATD identification and categorization performance compared to baseline methods.
Authors: Oleg Savchenko, Florian List, Guillermo Franco Abell\'an, Noemi Anau Montel, Christoph Weniger
Abstract: Reconstructing cosmological initial conditions (ICs) from late-time observations is a difficult task, which relies on the use of computationally expensive simulators alongside sophisticated statistical methods to navigate multi-million dimensional parameter spaces. We present a simple method for Bayesian field reconstruction based on modeling the posterior distribution of the initial matter density field to be diagonal Gaussian in Fourier space, with its covariance and the mean estimator being the trainable parts of the algorithm. Training and sampling are extremely fast (training: $\sim 1 \, \mathrm{h}$ on a GPU, sampling: $\lesssim 3 \, \mathrm{s}$ for 1000 samples at resolution $128^3$), and our method supports industry-standard (non-differentiable) $N$-body simulators. We verify the fidelity of the obtained IC samples in terms of summary statistics.
Authors: B\'alint M\'at\'e, Fran\c{c}ois Fleuret, Tristan Bereau
Abstract: We propose to compute solvation free energies via thermodynamic integration along a neural-network potential interpolating between two target Hamiltonians. We use a stochastic interpolant to define an interpolation between the distributions at the level of samples and optimize a neural network potential to match the corresponding equilibrium potential at every intermediate time-step. Once the alignment between the interpolating samples and the interpolating potentials is sufficiently accurate, the free-energy difference between the two Hamiltonians can be estimated using (neural) thermodynamic integration. We validate our method to compute solvation free energies on several benchmark systems: a Lennard-Jones particle in a Lennard-Jones fluid, as well as the insertion of both water and methane solutes in a water solvent at atomistic resolution.
Authors: Anika Hannemann, Arjhun Swaminathan, Ali Burak \"Unal, Mete Akg\"un
Abstract: Medical imaging is key in modern medicine. From magnetic resonance imaging (MRI) to microscopic imaging for blood cell detection, diagnostic medical imaging reveals vital insights into patient health. To predict diseases or provide individualized therapies, machine learning techniques like kernel methods have been widely used. Nevertheless, there are multiple challenges for implementing kernel methods. Medical image data often originates from various hospitals and cannot be combined due to privacy concerns, and the high dimensionality of image data presents another significant obstacle. While randomised encoding offers a promising direction, existing methods often struggle with a trade-off between accuracy and efficiency. Addressing the need for efficient privacy-preserving methods on distributed image data, we introduce OKRA (Orthonormal K-fRAmes), a novel randomized encoding-based approach for kernel-based machine learning. This technique, tailored for widely used kernel functions, significantly enhances scalability and speed compared to current state-of-the-art solutions. Through experiments conducted on various clinical image datasets, we evaluated model quality, computational performance, and resource overhead. Additionally, our method outperforms comparable approaches
Authors: Tailai Song, Paolo Garza, Michela Meo, Maurizio Matteo Munaf\`o
Abstract: The Real-time Transport Protocol (RTP)-based real-time communications (RTC) applications, exemplified by video conferencing, have experienced an unparalleled surge in popularity and development in recent years. In pursuit of optimizing their performance, the prediction of Quality of Service (QoS) metrics emerges as a pivotal endeavor, bolstering network monitoring and proactive solutions. However, contemporary approaches are confined to individual RTP flows and metrics, falling short in relationship capture and computational efficiency. To this end, we propose Packet-to-Prediction (P2P), a novel deep learning (DL) framework that hinges on raw packets to simultaneously process concurrent RTP flows and perform end-to-end prediction of multiple QoS metrics. Specifically, we implement a streamlined architecture, namely length-free Transformer with cross and neighbourhood attention, capable of handling an unlimited number of RTP flows, and employ a multi-task learning paradigm to forecast four key metrics in a single shot. Our work is based on extensive traffic collected during real video calls, and conclusively, P2P excels comparative models in both prediction performance and temporal efficiency.
Authors: Jingyu Guo, Christos Matsoukas, Fredrik Strand, Kevin Smith
Abstract: In multi-view medical diagnosis, deep learning-based models often fuse information from different imaging perspectives to improve diagnostic performance. However, existing approaches are prone to overfitting and rely heavily on view-specific features, which can lead to trivial solutions. In this work, we introduce Random Token Fusion (RTF), a novel technique designed to enhance multi-view medical image analysis using vision transformers. By integrating randomness into the feature fusion process during training, RTF addresses the issue of overfitting and enhances the robustness and accuracy of diagnostic models without incurring any additional cost at inference. We validate our approach on standard mammography and chest X-ray benchmark datasets. Through extensive experiments, we demonstrate that RTF consistently improves the performance of existing fusion methods, paving the way for a new generation of multi-view medical foundation models.
Authors: Sandeep Nagar, Mark Hasegawa-Johnson, David G. Beiser, Narendra Ahuja
Abstract: The COVID-19 pandemic has underscored the need for low-cost, scalable approaches to measuring contactless vital signs, either during initial triage at a healthcare facility or virtual telemedicine visits. Remote photoplethysmography (rPPG) can accurately estimate heart rate (HR) when applied to close-up videos of healthy volunteers in well-lit laboratory settings. However, results from such highly optimized laboratory studies may not be readily translated to healthcare settings. One significant barrier to the practical application of rPPG in health care is the accurate localization of the region of interest (ROI). Clinical or telemedicine visits may involve sub-optimal lighting, movement artifacts, variable camera angle, and subject distance. This paper presents an rPPG ROI selection method based on 3D facial landmarks and patient head yaw angle. We then demonstrate the robustness of this ROI selection method when coupled to the Plane-Orthogonal-to-Skin (POS) rPPG method when applied to videos of patients presenting to an Emergency Department for respiratory complaints. Our results demonstrate the effectiveness of our proposed approach in improving the accuracy and robustness of rPPG in a challenging clinical environment.
Authors: Hugh Greatorex, Ole Richter, Michele Mastella, Madison Cotteret, Philipp Klein, Maxime Fabre, Arianna Rubino, Willian Soares Gir\~ao, Junren Chen, Martin Ziegler, Laura B\'egon-Lours, Giacomo Indiveri, Elisabetta Chicca
Abstract: Recent advances in memory technologies, devices and materials have shown great potential for integration into neuromorphic electronic systems. However, a significant gap remains between the development of these materials and the realization of large-scale, fully functional systems. One key challenge is determining which devices and materials are best suited for specific functions and how they can be paired with CMOS circuitry. To address this, we introduce TEXEL, a mixed-signal neuromorphic architecture designed to explore the integration of on-chip learning circuits and novel two- and three-terminal devices. TEXEL serves as an accessible platform to bridge the gap between CMOS-based neuromorphic computation and the latest advancements in emerging devices. In this paper, we demonstrate the readiness of TEXEL for device integration through comprehensive chip measurements and simulations. TEXEL provides a practical system for testing bio-inspired learning algorithms alongside emerging devices, establishing a tangible link between brain-inspired computation and cutting-edge device research.
Authors: Bohdan M. Pavlyshenko
Abstract: The paper considers an approach of using Google Search API and GPT-4o model for qualitative and quantitative analyses of news through retrieval-augmented generation (RAG). This approach was applied to analyze news about the 2024 US presidential election process. Different news sources for different time periods have been analyzed. Quantitative scores generated by GPT model have been analyzed using Bayesian regression to derive trend lines. The distributions found for the regression parameters allow for the analysis of uncertainty in the election process. The obtained results demonstrate that using the GPT models for news analysis, one can get informative analytics and provide key insights that can be applied in further analyses of election processes.
Authors: Yuncheng Yuan, P\'eter Scheepers, Lydia Tasiou, Yunus Can G\"ultekin, Federico Corradi, Alex Alvarado
Abstract: This paper analyzes the design and competitiveness of four neural network (NN) architectures recently proposed as decoders for forward error correction (FEC) codes. We first consider the so-called single-label neural network (SLNN) and the multi-label neural network (MLNN) decoders which have been reported to achieve near maximum likelihood (ML) performance. Here, we show analytically that SLNN and MLNN decoders can always achieve ML performance, regardless of the code dimensions -- although at the cost of computational complexity -- and no training is in fact required. We then turn our attention to two transformer-based decoders: the error correction code transformer (ECCT) and the cross-attention message passing transformer (CrossMPT). We compare their performance against traditional decoders, and show that ordered statistics decoding outperforms these transformer-based decoders. The results in this paper cast serious doubts on the application of NN-based FEC decoders in the short and medium block length regime.
Authors: Sheheryar Mehmood, Peter Ochs
Abstract: Numerous Optimization Algorithms have a time-varying update rule thanks to, for instance, a changing step size, momentum parameter or, Hessian approximation. In this paper, we apply unrolled or automatic differentiation to a time-varying iterative process and provide convergence (rate) guarantees for the resulting derivative iterates. We adapt these convergence results and apply them to proximal gradient descent with variable step size and FISTA when solving partly smooth problems. We confirm our findings numerically by solving $\ell_1$ and $\ell_2$-regularized linear and logisitc regression respectively. Our theoretical and numerical results show that the convergence rate of the algorithm is reflected in its derivative iterates.
Authors: Azmine Toushik Wasi, Taki Hasan Rafi, Raima Islam, Karlo Serbetar, Dong Kyu Chae
Abstract: Reliable facial expression learning (FEL) involves the effective learning of distinctive facial expression characteristics for more reliable, unbiased and accurate predictions in real-life settings. However, current systems struggle with FEL tasks because of the variance in people's facial expressions due to their unique facial structures, movements, tones, and demographics. Biased and imbalanced datasets compound this challenge, leading to wrong and biased prediction labels. To tackle these, we introduce GReFEL, leveraging Vision Transformers and a facial geometry-aware anchor-based reliability balancing module to combat imbalanced data distributions, bias, and uncertainty in facial expression learning. Integrating local and global data with anchors that learn different facial data points and structural features, our approach adjusts biased and mislabeled emotions caused by intra-class disparity, inter-class similarity, and scale sensitivity, resulting in comprehensive, accurate, and reliable facial expression predictions. Our model outperforms current state-of-the-art methodologies, as demonstrated by extensive experiments on various datasets.
Authors: Szymon Bobek, Paloma Koryci\'nska, Monika Krakowska, Maciej Mozolewski, Dorota Rak, Magdalena Zych, Magdalena W\'ojcik, Grzegorz J. Nalepa
Abstract: This study is located in the Human-Centered Artificial Intelligence (HCAI) and focuses on the results of a user-centered assessment of commonly used eXplainable Artificial Intelligence (XAI) algorithms, specifically investigating how humans understand and interact with the explanations provided by these algorithms. To achieve this, we employed a multi-disciplinary approach that included state-of-the-art research methods from social sciences to measure the comprehensibility of explanations generated by a state-of-the-art lachine learning model, specifically the Gradient Boosting Classifier (XGBClassifier). We conducted an extensive empirical user study involving interviews with 39 participants from three different groups, each with varying expertise in data science, data visualization, and domain-specific knowledge related to the dataset used for training the machine learning model. Participants were asked a series of questions to assess their understanding of the model's explanations. To ensure replicability, we built the model using a publicly available dataset from the UC Irvine Machine Learning Repository, focusing on edible and non-edible mushrooms. Our findings reveal limitations in existing XAI methods and confirm the need for new design principles and evaluation techniques that address the specific information needs and user perspectives of different classes of AI stakeholders. We believe that the results of our research and the cross-disciplinary methodology we developed can be successfully adapted to various data types and user profiles, thus promoting dialogue and address opportunities in HCAI research. To support this, we are making the data resulting from our study publicly available.
Authors: Ram Mohan Rao Kadiyala
Abstract: This paper presents a detailed system description of our entry for the WASSA 2024 Task 2, focused on cross-lingual emotion detection. We utilized a combination of large language models (LLMs) and their ensembles to effectively understand and categorize emotions across different languages. Our approach not only outperformed other submissions with a large margin, but also demonstrated the strength of integrating multiple models to enhance performance. Additionally, We conducted a thorough comparison of the benefits and limitations of each model used. An error analysis is included along with suggested areas for future improvement. This paper aims to offer a clear and comprehensive understanding of advanced techniques in emotion detection, making it accessible even to those new to the field.
Authors: Hongkuan Zhou, Lavdim Halilaj, Sebastian Monka, Stefan Schmid, Yuqicheng Zhu, Bo Xiong, Steffen Staab
Abstract: Despite the remarkable success of deep neural networks (DNNs) in computer vision, they fail to remain high-performing when facing distribution shifts between training and testing data. In this paper, we propose Knowledge-Guided Visual representation learning (KGV), a distribution-based learning approach leveraging multi-modal prior knowledge, to improve generalization under distribution shift. We use prior knowledge from two distinct modalities: 1) a knowledge graph (KG) with hierarchical and association relationships; and 2) generated synthetic images of visual elements semantically represented in the KG. The respective embeddings are generated from the given modalities in a common latent space, i.e., visual embeddings from original and synthetic images as well as knowledge graph embeddings (KGEs). These embeddings are aligned via a novel variant of translation-based KGE methods, where the node and relation embeddings of the KG are modeled as Gaussian distributions and translations respectively. We claim that incorporating multi-model prior knowledge enables more regularized learning of image representations. Thus, the models are able to better generalize across different data distributions. We evaluate KGV on different image classification tasks with major or minor distribution shifts, namely road sign classification across datasets from Germany, China, and Russia, image classification with the mini-ImageNet dataset and its variants, as well as the DVM-CAR dataset. The results demonstrate that KGV consistently exhibits higher accuracy and data efficiency than the baselines across all experiments.
Authors: Mohammad Farazmand
Abstract: Discrete Empirical Interpolation Method (DEIM) estimates a function from its pointwise incomplete observations. In particular, this method can be used to estimate the state of a dynamical system from observational data gathered by sensors. However, when the number of observations are limited, DEIM returns large estimation errors. Sparse DEIM (S-DEIM) was recently developed to address this problem by introducing a kernel vector which previous DEIM-based methods had ignored. Unfortunately, estimating the optimal kernel vector in S-DEIM is a difficult task. Here, we introduce a data-driven method to estimate this kernel vector from sparse observational time series using recurrent neural networks. Using numerical examples, we demonstrate that this machine learning approach together with S-DEIM leads to nearly optimal state estimations.
Authors: Morenikeji Neri, Thomas Powell
Abstract: The Robbins-Siegmund theorem is one of the most important results in stochastic optimization, where it is widely used to prove the convergence of stochastic algorithms. We provide a quantitative version of the theorem, establishing a bound on how far one needs to look in order to locate a region of metastability in the sense of Tao. Our proof involves a metastable analogue of Doob's theorem for $L_1$-supermartingales along with a series of technical lemmas that make precise how quantitative information propagates through sums and products of stochastic processes. In this way, our paper establishes a general methodology for finding metastable bounds for stochastic processes that can be reduced to supermartingales, and therefore for obtaining quantitative convergence information across a broad class of stochastic algorithms whose convergence proof relies on some variation of the Robbins-Siegmund theorem. We conclude by discussing how our general quantitative result might be used in practice.
Authors: Matthias Bitzer, Reinis Cimurs, Benjamin Coors, Johannes Goth, Sebastian Ziesche, Philipp Geiger, Maximilian Naumann
Abstract: Simulation plays a crucial role in the rapid development and safe deployment of autonomous vehicles. Realistic traffic agent models are indispensable for bridging the gap between simulation and the real world. Many existing approaches for imitating human behavior are based on learning from demonstration. However, these approaches are often constrained by focusing on individual training strategies. Therefore, to foster a broader understanding of realistic traffic agent modeling, in this paper, we provide an extensive comparative analysis of different training principles, with a focus on closed-loop methods for highway driving simulation. We experimentally compare (i) open-loop vs. closed-loop multi-agent training, (ii) adversarial vs. deterministic supervised training, (iii) the impact of reinforcement losses, and (iv) the impact of training alongside log-replayed agents to identify suitable training techniques for realistic agent modeling. Furthermore, we identify promising combinations of different closed-loop training methods.
Authors: Ram Mohan Rao Kadiyala, Siddartha Pullakhandam, Kanwal Mehreen, Subhasya Tippareddy, Ashay Srivastava
Abstract: This paper presents our system description and error analysis of our entry for NLLP 2024 shared task on Legal Natural Language Inference (L-NLI) \citep{hagag2024legallenssharedtask2024}. The task required classifying these relationships as entailed, contradicted, or neutral, indicating any association between the review and the complaint. Our system emerged as the winning submission, significantly outperforming other entries with a substantial margin and demonstrating the effectiveness of our approach in legal text analysis. We provide a detailed analysis of the strengths and limitations of each model and approach tested, along with a thorough error analysis and suggestions for future improvements. This paper aims to contribute to the growing field of legal NLP by offering insights into advanced techniques for natural language inference in legal contexts, making it accessible to both experts and newcomers in the field.
Authors: Ram Mohan Rao Kadiyala, M. V. P. Chandra Sekhara Rao
Abstract: Social media is a great source of data for users reporting information and regarding their health and how various things have had an effect on them. This paper presents various approaches using Transformers and Large Language Models and their ensembles, their performance along with advantages and drawbacks for various tasks of SMM4H'24 - Classifying texts on impact of nature and outdoor spaces on the author's mental health (Task 3), Binary classification of tweets reporting their children's health disorders like Asthma, Autism, ADHD and Speech disorder (task 5), Binary classification of users self-reporting their age (task 6).
Authors: Seyed Hamed Haghshenas, Mia Naeini
Abstract: State Estimation is a crucial task in power systems. Graph Neural Networks have demonstrated significant potential in state estimation for power systems by effectively analyzing measurement data and capturing the complex interactions and interrelations among the measurements through the system's graph structure. However, the information about the system's graph structure may be inaccurate due to noise, attack or lack of accurate information about the topology of the system. This paper studies these scenarios under topology uncertainties and evaluates the impact of the topology uncertainties on the performance of a Temporal Graph Convolutional Network (TGCN) for state estimation in power systems. In order to make the model resilient to topology uncertainties, modifications in the TGCN model are proposed to incorporate a knowledge graph, generated based on the measurement data. This knowledge graph supports the assumed uncertain system graph. Two variations of the TGCN architecture are introduced to integrate the knowledge graph, and their performances are evaluated and compared to demonstrate improved resilience against topology uncertainties. The evaluation results indicate that while the two proposed architecture show different performance, they both improve the performance of the TGCN state estimation under topology uncertainties.
Authors: Abhijeet Kumar, Unnati Singh, Rajdeep Chatterjee, Tathagata Bandyopadhyay
Abstract: An efficient system of a queue control and regulation in public spaces is very important in order to avoid the traffic jams and to improve the customer satisfaction. This article offers a detailed road map based on a merger of intelligent systems and creating an efficient systems of queues in public places. Through the utilization of different technologies i.e. computer vision, machine learning algorithms, deep learning our system provide accurate information about the place is crowded or not and the necessary efforts to be taken.
Authors: Jiahao Qiu, Yifu Lu, Yifan Zeng, Jiacheng Guo, Jiayi Geng, Huazheng Wang, Kaixuan Huang, Yue Wu, Mengdi Wang
Abstract: Inference-time alignment enhances the performance of large language models without requiring additional training or fine-tuning but presents challenges due to balancing computational efficiency with high-quality output. Best-of-N (BoN) sampling, as a simple yet powerful approach, generates multiple responses and selects the best one, achieving improved performance but with a high computational cost. We propose TreeBoN, a novel framework that integrates a speculative tree-search strategy into Best-of-N (BoN) Sampling. TreeBoN maintains a set of parent nodes, iteratively branching and pruning low-quality responses, thereby reducing computational overhead while maintaining high output quality. Our approach also leverages token-level rewards from Direct Preference Optimization (DPO) to guide tree expansion and prune low-quality paths. We evaluate TreeBoN using AlpacaFarm, UltraFeedback, GSM8K, HH-RLHF, and TutorEval datasets, demonstrating consistent improvements. Specifically, TreeBoN achieves a 65% win rate at maximum lengths of 192 and 384 tokens, outperforming standard BoN with the same computational cost. Furthermore, TreeBoN achieves around a 60% win rate across longer responses, showcasing its scalability and alignment efficacy.
Authors: Isaac L. Huidobro-Meezs, Jun Dai, Guillaume Rabusseau, Rodrigo A. Vargas-Hern\'andez
Abstract: Quantum computing presents a promising alternative for the direct simulation of quantum systems with the potential to explore chemical problems beyond the capabilities of classical methods. However, current quantum algorithms are constrained by hardware limitations and the increased number of measurements required to achieve chemical accuracy. To address the measurement challenge, techniques for grouping commuting and anti-commuting terms, driven by heuristics, have been developed to reduce the number of measurements needed in quantum algorithms on near-term quantum devices. In this work, we propose a probabilistic framework using GFlowNets to group fully (FC) or qubit-wise commuting (QWC) terms within a given Hamiltonian. The significance of this approach is demonstrated by the reduced number of measurements for the found groupings; 51% and 67% reduction factors respectively for FC and QWC partitionings with respect to greedy coloring algorithms, highlighting the potential of GFlowNets for future applications in the measurement problem. Furthermore, the flexibility of our algorithm extends its applicability to other resource optimization problems in Hamiltonian simulation, such as circuit design.
Authors: Matteo Vilucchio, Nikolaos Tsilivis, Bruno Loureiro, Julia Kempe
Abstract: Regularization, whether explicit in terms of a penalty in the loss or implicit in the choice of algorithm, is a cornerstone of modern machine learning. Indeed, controlling the complexity of the model class is particularly important when data is scarce, noisy or contaminated, as it translates a statistical belief on the underlying structure of the data. This work investigates the question of how to choose the regularization norm $\lVert \cdot \rVert$ in the context of high-dimensional adversarial training for binary classification. To this end, we first derive an exact asymptotic description of the robust, regularized empirical risk minimizer for various types of adversarial attacks and regularization norms (including non-$\ell_p$ norms). We complement this analysis with a uniform convergence analysis, deriving bounds on the Rademacher Complexity for this class of problems. Leveraging our theoretical results, we quantitatively characterize the relationship between perturbation size and the optimal choice of $\lVert \cdot \rVert$, confirming the intuition that, in the data scarce regime, the type of regularization becomes increasingly important for adversarial training as perturbations grow in size.
Authors: Weichen Wu, Gen Li, Yuting Wei, Alessandro Rinaldo
Abstract: Statistical inference with finite-sample validity for the value function of a given policy in Markov decision processes (MDPs) is crucial for ensuring the reliability of reinforcement learning. Temporal Difference (TD) learning, arguably the most widely used algorithm for policy evaluation, serves as a natural framework for this purpose.In this paper, we study the consistency properties of TD learning with Polyak-Ruppert averaging and linear function approximation, and obtain three significant improvements over existing results. First, we derive a novel sharp high-dimensional probability convergence guarantee that depends explicitly on the asymptotic variance and holds under weak conditions. We further establish refined high-dimensional Berry-Esseen bounds over the class of convex sets that guarantee faster rates than those in the literature. Finally, we propose a plug-in estimator for the asymptotic covariance matrix, designed for efficient online computation. These results enable the construction of confidence regions and simultaneous confidence intervals for the linear parameters of the value function, with guaranteed finite-sample coverage. We demonstrate the applicability of our theoretical findings through numerical experiments.
Authors: Matthieu Haeberle, Puck van Gerwen, Ruben Laplaza, Ksenia R. Briling, Jan Weinreich, Friedrich Eisenbrand, Clemence Corminboeuf
Abstract: Integer linear programming (ILP) is an elegant approach to solve linear optimization problems, naturally described using integer decision variables. Within the context of physics-inspired machine learning applied to chemistry, we demonstrate the relevance of an ILP formulation to select molecular training sets for predictions of size-extensive properties. We show that our algorithm outperforms existing unsupervised training set selection approaches, especially when predicting properties of molecules larger than those present in the training set. We argue that the reason for the improved performance is due to the selection that is based on the notion of local similarity (i.e., per-atom) and a unique ILP approach that finds optimal solutions efficiently. Altogether, this work provides a practical algorithm to improve the performance of physics-inspired machine learning models and offers insights into the conceptual differences with existing training set selection approaches.
Authors: Rongxing Liu, Kumar Shridhar, Manish Prajapat, Patrick Xia, Mrinmaya Sachan
Abstract: Tasks requiring deductive reasoning, especially those involving multiple steps, often demand adaptive strategies such as intermediate generation of rationales or programs, as no single approach is universally optimal. While Language Models (LMs) can enhance their outputs through iterative self-refinement and strategy adjustments, they frequently fail to apply the most effective strategy in their first attempt. This inefficiency raises the question: Can LMs learn to select the optimal strategy in the first attempt, without a need for refinement? To address this challenge, we introduce SMART (Self-learning Meta-strategy Agent for Reasoning Tasks), a novel framework that enables LMs to autonomously learn and select the most effective strategies for various reasoning tasks. We model the strategy selection process as a Markov Decision Process and leverage reinforcement learning-driven continuous self-improvement to allow the model to find the suitable strategy to solve a given task. Unlike traditional self-refinement methods that rely on multiple inference passes or external feedback, SMART allows an LM to internalize the outcomes of its own reasoning processes and adjust its strategy accordingly, aiming for correct solutions on the first attempt. Our experiments across various reasoning datasets and with different model architectures demonstrate that SMART significantly enhances the ability of models to choose optimal strategies without external guidance (+15 points on the GSM8K dataset). By achieving higher accuracy with a single inference pass, SMART not only improves performance but also reduces computational costs for refinement-based strategies, paving the way for more efficient and intelligent reasoning in LMs.
Authors: Giannis Daras, Weili Nie, Karsten Kreis, Alex Dimakis, Morteza Mardani, Nikola Borislavov Kovachki, Arash Vahdat
Abstract: Using image models naively for solving inverse video problems often suffers from flickering, texture-sticking, and temporal inconsistency in generated videos. To tackle these problems, in this paper, we view frames as continuous functions in the 2D space, and videos as a sequence of continuous warping transformations between different frames. This perspective allows us to train function space diffusion models only on images and utilize them to solve temporally correlated inverse problems. The function space diffusion models need to be equivariant with respect to the underlying spatial transformations. To ensure temporal consistency, we introduce a simple post-hoc test-time guidance towards (self)-equivariant solutions. Our method allows us to deploy state-of-the-art latent diffusion models such as Stable Diffusion XL to solve video inverse problems. We demonstrate the effectiveness of our method for video inpainting and $8\times$ video super-resolution, outperforming existing techniques based on noise transformations. We provide generated video results: https://giannisdaras.github.io/warped\_diffusion.github.io/.
URLs: https://giannisdaras.github.io/warped\_diffusion.github.io/.
Authors: Alexander Bienstock, Ujjwal Kumar, Antigoni Polychroniadou
Abstract: Federated Learning (FL) has gained lots of traction recently, both in industry and academia. In FL, a machine learning model is trained using data from various end-users arranged in committees across several rounds. Since such data can often be sensitive, a primary challenge in FL is providing privacy while still retaining utility of the model. Differential Privacy (DP) has become the main measure of privacy in the FL setting. DP comes in two flavors: central and local. In the former, a centralized server is trusted to receive the users' raw gradients from a training step, and then perturb their aggregation with some noise before releasing the next version of the model. In the latter (more private) setting, noise is applied on users' local devices, and only the aggregation of users' noisy gradients is revealed even to the server. Great strides have been made in increasing the privacy-utility trade-off in the central DP setting, by utilizing the so-called matrix mechanism. However, progress has been mostly stalled in the local DP setting. In this work, we introduce the distributed matrix mechanism to achieve the best-of-both-worlds; local DP and also better privacy-utility trade-off from the matrix mechanism. We accomplish this by proposing a cryptographic protocol that securely transfers sensitive values across rounds, which makes use of packed secret sharing. This protocol accommodates the dynamic participation of users per training round required by FL, including those that may drop out from the computation. We provide experiments which show that our mechanism indeed significantly improves the privacy-utility trade-off of FL models compared to previous local DP mechanisms, with little added overhead.
Authors: Zhuoming Chen, Ranajoy Sadhukhan, Zihao Ye, Yang Zhou, Jianyu Zhang, Niklas Nolte, Yuandong Tian, Matthijs Douze, Leon Bottou, Zhihao Jia, Beidi Chen
Abstract: Large language models (LLMs) with long context windows have gained significant attention. However, the KV cache, stored to avoid re-computation, becomes a bottleneck. Various dynamic sparse or TopK-based attention approximation methods have been proposed to leverage the common insight that attention is sparse. In this paper, we first show that TopK attention itself suffers from quality degradation in certain downstream tasks because attention is not always as sparse as expected. Rather than selecting the keys and values with the highest attention scores, sampling with theoretical guarantees can provide a better estimation for attention output. To make the sampling-based approximation practical in LLM generation, we propose MagicPIG, a heterogeneous system based on Locality Sensitive Hashing (LSH). MagicPIG significantly reduces the workload of attention computation while preserving high accuracy for diverse tasks. MagicPIG stores the LSH hash tables and runs the attention computation on the CPU, which allows it to serve longer contexts and larger batch sizes with high approximation accuracy. MagicPIG can improve decoding throughput by $1.9\sim3.9\times$ across various GPU hardware and achieve 110ms decoding latency on a single RTX 4090 for Llama-3.1-8B-Instruct model with a context of 96k tokens. The code is available at \url{https://github.com/Infini-AI-Lab/MagicPIG}.
Authors: Niclas Dern, John P. Cunningham, Geoff Pleiss
Abstract: Classic tree-based ensembles generalize better than any single decision tree. In contrast, recent empirical studies find that modern ensembles of (overparameterized) neural networks may not provide any inherent generalization advantage over single but larger neural networks. This paper clarifies how modern overparameterized ensembles differ from their classic underparameterized counterparts, using ensembles of random feature (RF) regressors as a basis for developing theory. In contrast to the underparameterized regime, where ensembling typically induces regularization and increases generalization, we prove that infinite ensembles of overparameterized RF regressors become pointwise equivalent to (single) infinite-width RF regressors. This equivalence, which is exact for ridgeless models and approximate for small ridge penalties, implies that overparameterized ensembles and single large models exhibit nearly identical generalization. As a consequence, we can characterize the predictive variance amongst ensemble members, and demonstrate that it quantifies the expected effects of increasing capacity rather than capturing any conventional notion of uncertainty. Our results challenge common assumptions about the advantages of ensembles in overparameterized settings, prompting a reconsideration of how well intuitions from underparameterized ensembles transfer to deep ensembles and the overparameterized regime.
Authors: Kumar Manas, Stefan Zwicklbauer, Adrian Paschke
Abstract: Autonomous agents often face the challenge of interpreting uncertain natural language instructions for planning tasks. Representing these instructions as Linear Temporal Logic (LTL) enables planners to synthesize actionable plans. We introduce CoT-TL, a data-efficient in-context learning framework for translating natural language specifications into LTL representations. CoT-TL addresses the limitations of large language models, which typically rely on extensive fine-tuning data, by extending chain-of-thought reasoning and semantic roles to align with the requirements of formal logic creation. This approach enhances the transparency and rationale behind LTL generation, fostering user trust. CoT-TL achieves state-of-the-art accuracy across three diverse datasets in low-data scenarios, outperforming existing methods without fine-tuning or intermediate translations. To improve reliability and minimize hallucinations, we incorporate model checking to validate the syntax of the generated LTL output. We further demonstrate CoT-TL's effectiveness through ablation studies and evaluations on unseen LTL structures and formulas in a new dataset. Finally, we validate CoT-TL's practicality by integrating it into a QuadCopter for multi-step drone planning based on natural language instructions.
Authors: L. I. Zablocki, L. A. Bugnon, M. Gerard, L. Di Persia, G. Stegmayer, D. H. Milone
Abstract: Inspired by the success of large language models (LLM) for DNA and proteins, several LLM for RNA have been developed recently. RNA-LLM uses large datasets of RNA sequences to learn, in a self-supervised way, how to represent each RNA base with a semantically rich numerical vector. This is done under the hypothesis that obtaining high-quality RNA representations can enhance data-costly downstream tasks. Among them, predicting the secondary structure is a fundamental task for uncovering RNA functional mechanisms. In this work we present a comprehensive experimental analysis of several pre-trained RNA-LLM, comparing them for the RNA secondary structure prediction task in an unified deep learning framework. The RNA-LLM were assessed with increasing generalization difficulty on benchmark datasets. Results showed that two LLM clearly outperform the other models, and revealed significant challenges for generalization in low-homology scenarios.
Authors: Samrajya Thapa, Koushik Howlader, Subhankar Bhattacharjee, Wei le
Abstract: In this paper, we introduce a novel Multi-Modal Contrastive Pre-training Framework that synergistically combines X-rays, electrocardiograms (ECGs), and radiology/cardiology reports. Our approach leverages transformers to encode these diverse modalities into a unified representation space, aiming to enhance diagnostic accuracy and facilitate comprehensive patient assessments. We utilize LoRA-Peft to significantly reduce trainable parameters in the LLM and incorporate recent linear attention dropping strategy in the Vision Transformer(ViT) for smoother attention. Furthermore, we provide novel multimodal attention explanations and retrieval for our model. To the best of our knowledge, we are the first to propose an integrated model that combines X-ray, ECG, and Radiology/Cardiology Report with this approach. By utilizing contrastive loss, MoRE effectively aligns modality-specific features into a coherent embedding, which supports various downstream tasks such as zero-shot classification and multimodal retrieval. Employing our proposed methodology, we achieve state-of-the-art (SOTA) on the Mimic-IV, CheXpert, Edema Severity, and PtbXl downstream datasets, surpassing existing multimodal approaches. Our proposed framework shows significant improvements in capturing intricate inter-modal relationships and its robustness in medical diagnosis that establishes a framework for future research in multimodal learning in the healthcare sector.
Authors: Sukanya Patra, Souhaib Ben Taieb
Abstract: Industrial anomaly detection is crucial for quality control and predictive maintenance, but it presents challenges due to limited training data, diverse anomaly types, and external factors that alter object appearances. Existing methods commonly detect structural anomalies, such as dents and scratches, by leveraging multi-scale features from image patches extracted through deep pre-trained networks. However, significant memory and computational demands often limit their practical application. Additionally, detecting logical anomalies-such as images with missing or excess elements-requires an understanding of spatial relationships that traditional patch-based methods fail to capture. In this work, we address these limitations by focusing on Deep Feature Reconstruction (DFR), a memory- and compute-efficient approach for detecting structural anomalies. We further enhance DFR into a unified framework, called ULSAD, which is capable of detecting both structural and logical anomalies. Specifically, we refine the DFR training objective to improve performance in structural anomaly detection, while introducing an attention-based loss mechanism using a global autoencoder-like network to handle logical anomaly detection. Our empirical evaluation across five benchmark datasets demonstrates the performance of ULSAD in detecting and localizing both structural and logical anomalies, outperforming eight state-of-the-art methods. An extensive ablation study further highlights the contribution of each component to the overall performance improvement. Our code is available at https://github.com/sukanyapatra1997/ULSAD-2024.git
Authors: Michael S. Ryoo, Honglu Zhou, Shrikant Kendre, Can Qin, Le Xue, Manli Shu, Silvio Savarese, Ran Xu, Caiming Xiong, Juan Carlos Niebles
Abstract: We present xGen-MM-Vid (BLIP-3-Video): a multimodal language model for videos, particularly designed to efficiently capture temporal information over multiple frames. BLIP-3-Video takes advantage of the 'temporal encoder' in addition to the conventional visual tokenizer, which maps a sequence of tokens over multiple frames into a compact set of visual tokens. This enables BLIP3-Video to use much fewer visual tokens than its competing models (e.g., 32 vs. 4608 tokens). We explore different types of temporal encoders, including learnable spatio-temporal pooling as well as sequential models like Token Turing Machines. We experimentally confirm that BLIP-3-Video obtains video question-answering accuracies comparable to much larger state-of-the-art models (e.g., 34B), while being much smaller (i.e., 4B) and more efficient by using fewer visual tokens. The project website is at https://www.salesforceairesearch.com/opensource/xGen-MM-Vid/index.html
URLs: https://www.salesforceairesearch.com/opensource/xGen-MM-Vid/index.html
Authors: Jad Salem, Swati Gupta, Vijay Kamble
Abstract: Algorithmic decision-making in societal contexts, such as retail pricing, loan administration, recommendations on online platforms, etc., can be framed as stochastic optimization under bandit feedback, which typically requires experimentation with different decisions for the sake of learning. Such experimentation often results in perceptions of unfairness among people impacted by these decisions; for instance, there have been several recent lawsuits accusing companies that deploy algorithmic pricing practices of price gouging. Motivated by the changing legal landscape surrounding algorithmic decision-making, we introduce the well-studied fairness notion of envy-freeness within the context of stochastic convex optimization. Our notion requires that upon receiving decisions in the present time, groups do not envy the decisions received by any of the other groups, both in the present as well as the past. This results in a novel trajectory-constrained stochastic optimization problem that renders existing techniques inapplicable. The main technical contribution of this work is to show problem settings where there is no gap in achievable regret (up to logarithmic factors) when envy-freeness is imposed. In particular, in our main result, we develop a near-optimal envy-free algorithm that achieves $\tilde{O}(\sqrt{T})$ regret for smooth convex functions that satisfy the PL inequality. This algorithm has a coordinate-descent structure, in which we carefully leverage gradient information to ensure monotonic sampling along each dimension, while avoiding overshooting the constrained optimum with high probability. This latter aspect critically uses smoothness and the structure of the envy-freeness constraints, while the PL inequality allows for sufficient progress towards the optimal solution. We discuss several open questions that arise from this analysis, which may be of independent interest.
Authors: Laurent Orseau, Marcus Hutter
Abstract: We extend and combine several tools of the literature to design fast, adaptive, anytime and scale-free online learning algorithms. Scale-free regret bounds must scale linearly with the maximum loss, both toward large losses and toward very small losses. Adaptive regret bounds demonstrate that an algorithm can take advantage of easy data and potentially have constant regret. We seek to develop fast algorithms that depend on as few parameters as possible, in particular they should be anytime and thus not depend on the time horizon. Our first and main tool, isotuning, is a generalization of the idea of designing adaptive learning rates that balance the trade-off of the regret. We provide a simple and versatile theorem that can be applied to a wide range of settings, and competes with the best balancing in hindsight within a factor 2. The second tool is an online correction, which allows us to obtain centered bounds for many algorithms, to prevent the regret bounds from being vacuous when the domain is overly large or only partially constrained. The last tool, null updates, prevents the algorithm from performing overly large updates, which could result in unbounded regret, or even invalid updates. We develop a general theory to combine all these tools and apply it to several standard algorithms. In particular, we (almost entirely) restore the adaptivity to small losses of FTRL for unbounded domains, design and prove scale-free adaptive guarantees for a variant of Mirror Descent (at least when the Bregman divergence is convex in its second argument), extend Adapt-ML-Prod to scale-free guarantees, and provide several additional contributions about Prod, AdaHedge, BOA and Soft-Bayes.
Authors: Sajad Khodadadian, Pranay Sharma, Gauri Joshi, Siva Theja Maguluri
Abstract: Since reinforcement learning algorithms are notoriously data-intensive, the task of sampling observations from the environment is usually split across multiple agents. However, transferring these observations from the agents to a central location can be prohibitively expensive in terms of communication cost, and it can also compromise the privacy of each agent's local behavior policy. Federated reinforcement learning is a framework in which $N$ agents collaboratively learn a global model, without sharing their individual data and policies. This global model is the unique fixed point of the average of $N$ local operators, corresponding to the $N$ agents. Each agent maintains a local copy of the global model and updates it using locally sampled data. In this paper, we show that by careful collaboration of the agents in solving this joint fixed point problem, we can find the global model $N$ times faster, also known as linear speedup. We first propose a general framework for federated stochastic approximation with Markovian noise and heterogeneity, showing linear speedup in convergence. We then apply this framework to federated reinforcement learning algorithms, examining the convergence of federated on-policy TD, off-policy TD, and $Q$-learning.
Authors: Xiangming Meng, Yoshiyuki Kabashima
Abstract: With the rapid development of diffusion models and flow-based generative models, there has been a surge of interests in solving noisy linear inverse problems, e.g., super-resolution, deblurring, denoising, colorization, etc, with generative models. However, while remarkable reconstruction performances have been achieved, their inference time is typically too slow since most of them rely on the seminal diffusion posterior sampling (DPS) framework and thus to approximate the intractable likelihood score, time-consuming gradient calculation through back-propagation is needed. To address this issue, this paper provides a fast and effective solution by proposing a simple closed-form approximation to the likelihood score. For both diffusion and flow-based models, extensive experiments are conducted on various noisy linear inverse problems such as noisy super-resolution, denoising, deblurring, and colorization. In all these tasks, our method (namely DMPS) demonstrates highly competitive or even better reconstruction performances while being significantly faster than all the baseline methods.
Authors: Paul Hagemann, Sophie Mildenberger, Lars Ruthotto, Gabriele Steidl, Nicole Tianjiao Yang
Abstract: Score-based diffusion models (SBDM) have recently emerged as state-of-the-art approaches for image generation. Existing SBDMs are typically formulated in a finite-dimensional setting, where images are considered as tensors of finite size. This paper develops SBDMs in the infinite-dimensional setting, that is, we model the training data as functions supported on a rectangular domain. In addition to the quest for generating images at ever-higher resolutions, our primary motivation is to create a well-posed infinite-dimensional learning problem that we can discretize consistently on multiple resolution levels. We thereby intend to obtain diffusion models that generalize across different resolution levels and improve the efficiency of the training process. We demonstrate how to overcome two shortcomings of current SBDM approaches in the infinite-dimensional setting. First, we modify the forward process using trace class operators to ensure that the latent distribution is well-defined in the infinite-dimensional setting and derive the reverse processes for finite-dimensional approximations. Second, we illustrate that approximating the score function with an operator network is beneficial for multilevel training. After deriving the convergence of the discretization and the approximation of multilevel training, we demonstrate some practical benefits of our infinite-dimensional SBDM approach on a synthetic Gaussian mixture example, the MNIST dataset, and a dataset generated from a nonlinear 2D reaction-diffusion equation.
Authors: Hamid El Bahja, Jan Christian Hauffen, Peter Jung, Bubacarr Bah, Issa Karambal
Abstract: Deep learning has been highly successful in some applications. Nevertheless, its use for solving partial differential equations (PDEs) has only been of recent interest with current state-of-the-art machine learning libraries, e.g., TensorFlow or PyTorch. Physics-informed neural networks (PINNs) are an attractive tool for solving partial differential equations based on sparse and noisy data. Here extend PINNs to solve obstacle-related PDEs which present a great computational challenge because they necessitate numerical methods that can yield an accurate approximation of the solution that lies above a given obstacle. The performance of the proposed PINNs is demonstrated in multiple scenarios for linear and nonlinear PDEs subject to regular and irregular obstacles.
Authors: Gyuhak Kim, Changnan Xiao, Tatsuya Konishi, Zixuan Ke, Bing Liu
Abstract: As AI agents are increasingly used in the real open world with unknowns or novelties, they need the ability to (1) recognize objects that (a) they have learned before and (b) detect items that they have never seen or learned, and (2) learn the new items incrementally to become more and more knowledgeable and powerful. (1) is called novelty detection or out-of-distribution (OOD) detection and (2) is called class incremental learning (CIL), which is a setting of continual learning (CL). In existing research, OOD detection and CIL are regarded as two completely different problems. This paper first provides a theoretical proof that good OOD detection for each task within the set of learned tasks (called closed-world OOD detection) is necessary for successful CIL. We show this by decomposing CIL into two sub-problems: within-task prediction (WP) and task-id prediction (TP), and proving that TP is correlated with closed-world OOD detection. The key theoretical result is that regardless of whether WP and OOD detection (or TP) are defined explicitly or implicitly by a CIL algorithm, good WP and good closed-world OOD detection are necessary and sufficient conditions for good CIL, which unifies novelty or OOD detection and continual learning (CIL, in particular). We call this traditional CIL the closed-world CIL as it does not detect future OOD data in the open world. The paper then proves that the theory can be generalized or extended to open-world CIL, which is the proposed open-world continual learning, that can perform CIL in the open world and detect future or open-world OOD data. Based on the theoretical results, new CIL methods are also designed, which outperform strong baselines in CIL accuracy and in continual OOD detection by a large margin.
Authors: Daiwei Chen, Wei-Kai Chang, Pratik Chaudhari
Abstract: We use a formal correspondence between thermodynamics and inference, where the number of samples can be thought of as the inverse temperature, to study a quantity called ``learning capacity'' which is a measure of the effective dimensionality of a model. We show that the learning capacity is a useful notion of the complexity because (a) it correlates well with the test loss and it is a tiny fraction of the number of parameters for many deep networks trained on typical datasets, (b) it depends upon the number of samples used for training, (c) it is numerically consistent with notions of capacity obtained from PAC-Bayes generalization bounds, and (d) the test loss as a function of the learning capacity does not exhibit double descent. We show that the learning capacity saturates at very small and very large sample sizes; the threshold that characterizes the transition between these two regimes provides guidelines as to when one should procure more data and when one should search for a different architecture to improve performance. We show how the learning capacity can be used to provide a quantitative notion of capacity even for non-parametric models such as random forests and nearest neighbor classifiers.
Authors: Florentin Guth, Brice M\'enard, Gaspar Rochette, St\'ephane Mallat
Abstract: A central question in deep learning is to understand the functions learned by deep networks. What is their approximation class? Do the learned weights and representations depend on initialization? Previous empirical work has evidenced that kernels defined by network activations are similar across initializations. For shallow networks, this has been theoretically studied with random feature models, but an extension to deep networks has remained elusive. Here, we provide a deep extension of such random feature models, which we call the rainbow model. We prove that rainbow networks define deterministic (hierarchical) kernels in the infinite-width limit. The resulting functions thus belong to a data-dependent RKHS which does not depend on the weight randomness. We also verify numerically our modeling assumptions on deep CNNs trained on image classification tasks, and show that the trained networks approximately satisfy the rainbow hypothesis. In particular, rainbow networks sampled from the corresponding random feature model achieve similar performance as the trained networks. Our results highlight the central role played by the covariances of network weights at each layer, which are observed to be low-rank as a result of feature learning.
Authors: Cristina Improta, Pietro Liguori, Roberto Natella, Bojan Cukic, Domenico Cotroneo
Abstract: Since manually writing software exploits for offensive security is time-consuming and requires expert knowledge, AI-base code generators are an attractive solution to enhance security analysts' productivity by automatically crafting exploits for security testing. However, the variability in the natural language and technical skills used to describe offensive code poses unique challenges to their robustness and applicability. In this work, we present a method to add perturbations to the code descriptions to create new inputs in natural language (NL) from well-intentioned developers that diverge from the original ones due to the use of new words or because they miss part of them. The goal is to analyze how and to what extent perturbations affect the performance of AI code generators in the context of offensive code. First, we show that perturbed descriptions preserve the semantics of the original, non-perturbed ones. Then, we use the method to assess the robustness of three state-of-the-art code generators against the newly perturbed inputs, showing that the performance of these AI-based solutions is highly affected by perturbations in the NL descriptions. To enhance their robustness, we use the method to perform data augmentation, i.e., to increase the variability and diversity of the NL descriptions in the training data, proving its effectiveness against both perturbed and non-perturbed code descriptions.
Authors: Emmanuel Balogun, Ram Rajagopal, Arun Majumdar
Abstract: Stochastic generators are useful for estimating climate impacts on various sectors. Projecting climate risk in various sectors, e.g. energy systems, requires generators that are accurate (statistical resemblance to ground-truth), reliable (do not produce erroneous examples), and efficient. Leveraging data from the North American Land Data Assimilation System, we introduce TemperatureGAN, a Generative Adversarial Network conditioned on months, locations, and time periods, to generate 2m above ground atmospheric temperatures at an hourly resolution. We propose evaluation methods and metrics to measure the quality of generated samples. We show that TemperatureGAN produces high-fidelity examples with good spatial representation and temporal dynamics consistent with known diurnal cycles.
Authors: Xiang Liu, Liangxi Liu, Feiyang Ye, Yunheng Shen, Xia Li, Linshan Jiang, Jialin Li
Abstract: Efficiently aggregating trained neural networks from local clients into a global model on a server is a widely researched topic in federated learning. Recently, motivated by diminishing privacy concerns, mitigating potential attacks, and reducing communication overhead, one-shot federated learning (i.e., limiting client-server communication into a single round) has gained popularity among researchers. However, the one-shot aggregation performances are sensitively affected by the non-identical training data distribution, which exhibits high statistical heterogeneity in some real-world scenarios. To address this issue, we propose a novel one-shot aggregation method with layer-wise posterior aggregation, named FedLPA. FedLPA aggregates local models to obtain a more accurate global model without requiring extra auxiliary datasets or exposing any private label information, e.g., label distributions. To effectively capture the statistics maintained in the biased local datasets in the practical non-IID scenario, we efficiently infer the posteriors of each layer in each local model using layer-wise Laplace approximation and aggregate them to train the global parameters. Extensive experimental results demonstrate that FedLPA significantly improves learning performance over state-of-the-art methods across several metrics.
Authors: Ho-Joon Lee, Prashant S. Emani, Mark B. Gerstein
Abstract: The accurate screening of candidate drug ligands against target proteins through computational approaches is of prime interest to drug development efforts. Such virtual screening depends in part on methods to predict the binding affinity between ligands and proteins. Many computational models for binding affinity prediction have been developed, but with varying results across targets. Given that ensembling or meta-modeling approaches have shown great promise in reducing model-specific biases, we develop a framework to integrate published force-field-based empirical docking and sequence-based deep learning models. In building this framework, we evaluate many combinations of individual base models, training databases, and several meta-modeling approaches. We show that many of our meta-models significantly improve affinity predictions over base models. Our best meta-models achieve comparable performance to state-of-the-art deep learning tools exclusively based on 3D structures, while allowing for improved database scalability and flexibility through the explicit inclusion of features such as physicochemical properties or molecular descriptors. We further demonstrate improved generalization capability by our models using a large-scale benchmark of affinity prediction as well as a virtual screening application benchmark. Overall, we demonstrate that diverse modeling approaches can be ensembled together to gain meaningful improvement in binding affinity prediction.
Authors: Minxing Zhang, Michael Backes, Xiao Zhang
Abstract: This paper revisits the robust overfitting phenomenon of adversarial training. Observing that models with better robust generalization performance are less certain in predicting adversarially generated training inputs, we argue that overconfidence in predicting adversarial examples is a potential cause. Therefore, we hypothesize that generating less certain adversarial examples improves robust generalization, and propose a formal definition of adversarial certainty that captures the variance of the model's predicted logits on adversarial examples. Our theoretical analysis of synthetic distributions characterizes the connection between adversarial certainty and robust generalization. Accordingly, built upon the notion of adversarial certainty, we develop a general method to search for models that can generate training-time adversarial inputs with reduced certainty, while maintaining the model's capability in distinguishing adversarial examples. Extensive experiments on image benchmarks demonstrate that our method effectively learns models with consistently improved robustness and mitigates robust overfitting, confirming the importance of generating less certain adversarial examples for robust generalization.
Authors: Nizar Demni, Hachem Kadri
Abstract: Random features have been introduced to scale up kernel methods via randomization techniques. In particular, random Fourier features and orthogonal random features were used to approximate the popular Gaussian kernel. Random Fourier features are built in this case using a random Gaussian matrix. In this work, we analyze the bias and the variance of the kernel approximation based on orthogonal random features which makes use of Haar orthogonal matrices. We provide explicit expressions for these quantities using normalized Bessel functions, showing that orthogonal random features does not approximate the Gaussian kernel but a Bessel kernel. We also derive sharp exponential bounds supporting the view that orthogonal random features are less dispersed than random Fourier features.
Authors: Jiawen Zhang, Xumeng Wen, Zhenwei Zhang, Shun Zheng, Jia Li, Jiang Bian
Abstract: Delivering precise point and distributional forecasts across a spectrum of prediction horizons represents a significant and enduring challenge in the application of time-series forecasting within various industries. Prior research on developing deep learning models for time-series forecasting has often concentrated on isolated aspects, such as long-term point forecasting or short-term probabilistic estimations. This narrow focus may result in skewed methodological choices and hinder the adaptability of these models to uncharted scenarios. While there is a rising trend in developing universal forecasting models, a thorough understanding of their advantages and drawbacks, especially regarding essential forecasting needs like point and distributional forecasts across short and long horizons, is still lacking. In this paper, we present ProbTS, a benchmark tool designed as a unified platform to evaluate these fundamental forecasting needs and to conduct a rigorous comparative analysis of numerous cutting-edge studies from recent years. We dissect the distinctive data characteristics arising from disparate forecasting requirements and elucidate how these characteristics can skew methodological preferences in typical research trajectories, which often fail to fully accommodate essential forecasting needs. Building on this, we examine the latest models for universal time-series forecasting and discover that our analyses of methodological strengths and weaknesses are also applicable to these universal models. Finally, we outline the limitations inherent in current research and underscore several avenues for future exploration.
Authors: Lapo Frati, Neil Traft, Jeff Clune, Nick Cheney
Abstract: This work identifies a simple pre-training mechanism that leads to representations exhibiting better continual and transfer learning. This mechanism -- the repeated resetting of weights in the last layer, which we nickname "zapping" -- was originally designed for a meta-continual-learning procedure, yet we show it is surprisingly applicable in many settings beyond both meta-learning and continual learning. In our experiments, we wish to transfer a pre-trained image classifier to a new set of classes, in a few shots. We show that our zapping procedure results in improved transfer accuracy and/or more rapid adaptation in both standard fine-tuning and continual learning settings, while being simple to implement and computationally efficient. In many cases, we achieve performance on par with state of the art meta-learning without needing the expensive higher-order gradients, by using a combination of zapping and sequential learning. An intuitive explanation for the effectiveness of this zapping procedure is that representations trained with repeated zapping learn features that are capable of rapidly adapting to newly initialized classifiers. Such an approach may be considered a computationally cheaper type of, or alternative to, meta-learning rapidly adaptable features with higher-order gradients. This adds to recent work on the usefulness of resetting neural network parameters during training, and invites further investigation of this mechanism.
Authors: Khoa Tran, Lam Pham, Vy-Rin Nguyen, Ho-Si-Hung Nguyen
Abstract: Motor bearing fault detection (MBFD) is critical for maintaining the reliability and operational efficiency of industrial machinery. Early detection of bearing faults can prevent system failures, reduce operational downtime, and lower maintenance costs. In this paper, we propose a robust deep learning-based system for MBFD that incorporates multiple training strategies, including supervised, semi-supervised, and unsupervised learning. To enhance the detection performance, we introduce a novel double loss function. Our approach is evaluated using benchmark datasets from the American Society for Mechanical Failure Prevention Technology (MFPT), Case Western Reserve University Bearing Center (CWRU), and Paderborn University's Condition Monitoring of Bearing Damage in Electromechanical Drive Systems (PU). Results demonstrate that deep learning models outperform traditional machine learning techniques, with our novel system achieving superior accuracy across all datasets. These findings highlight the potential of our approach for practical MBFD applications.
Authors: Dai Hai Nguyen, Tetsuya Sakurai, Hiroshi Mamitsuka
Abstract: Variational inference (VI) can be cast as an optimization problem in which the variational parameters are tuned to closely align a variational distribution with the true posterior. The optimization task can be approached through vanilla gradient descent in black-box VI or natural-gradient descent in natural-gradient VI. In this work, we reframe VI as the optimization of an objective that concerns probability distributions defined over a \textit{variational parameter space}. Subsequently, we propose Wasserstein gradient descent for tackling this optimization problem. Notably, the optimization techniques, namely black-box VI and natural-gradient VI, can be reinterpreted as specific instances of the proposed Wasserstein gradient descent. To enhance the efficiency of optimization, we develop practical methods for numerically solving the discrete gradient flows. We validate the effectiveness of the proposed methods through empirical experiments on a synthetic dataset, supplemented by theoretical analyses.
Authors: Christian Jacobsen, Yilin Zhuang, Karthik Duraisamy
Abstract: Recent advances in generative artificial intelligence have had a significant impact on diverse domains spanning computer vision, natural language processing, and drug discovery. This work extends the reach of generative models into physical problem domains, particularly addressing the efficient enforcement of physical laws and conditioning for forward and inverse problems involving partial differential equations (PDEs). Our work introduces two key contributions: firstly, we present an efficient approach to promote consistency with the underlying PDE. By incorporating discretized information into score-based generative models, our method generates samples closely aligned with the true data distribution, showcasing residuals comparable to data generated through conventional PDE solvers, significantly enhancing fidelity. Secondly, we showcase the potential and versatility of score-based generative models in various physics tasks, specifically highlighting surrogate modeling as well as probabilistic field reconstruction and inversion from sparse measurements. A robust foundation is laid by designing unconditional score-based generative models that utilize reversible probability flow ordinary differential equations. Leveraging conditional models that require minimal training, we illustrate their flexibility when combined with a frozen unconditional model. These conditional models generate PDE solutions by incorporating parameters, macroscopic quantities, or partial field measurements as guidance. The results illustrate the inherent flexibility of score-based generative models and explore the synergy between unconditional score-based generative models and the present physically-consistent sampling approach, emphasizing the power and flexibility in solving for and inverting physical fields governed by differential equations, and in other scientific machine learning tasks.
Authors: Tianyi Li, Joshua Klavins, Te Xu, Niaz Mahmud Zafri, Raphael Stern
Abstract: Many factors influence the yielding result of a driver-pedestrian interaction, including traffic volume, vehicle speed, roadway characteristics, etc. While individual aspects of these interactions have been explored, comprehensive, naturalistic studies, particularly those considering the built environment's influence on driver-yielding behavior, are lacking. To address this gap, our study introduces an extensive open-source dataset, compiled from video data at 18 unsignalized intersections across Minnesota. Documenting more than 3000 interactions, this dataset provides a detailed view of driver-pedestrian interactions and over 50 distinct contextual variables. The data, which covers individual driver-pedestrian interactions and contextual factors, is made publicly available at https://github.com/tianyi17/pedestrian_yielding_data_MN. Using logistic regression, we developed a classification model that predicts driver yielding based on the identified variables. Our analysis indicates that vehicle speed, the presence of parking lots, proximity to parks or schools, and the width of major road crossings significantly influence driver yielding at unsignalized intersections. Through our findings and by publishing one of the most comprehensive driver-pedestrian datasets in the United States, our study will support communities across Minnesota and the United States in their ongoing efforts to improve road safety for pedestrians and be helpful for automated vehicle design.
URLs: https://github.com/tianyi17/pedestrian_yielding_data_MN.
Authors: Ye Lin Tun, Chu Myaet Thwal, Le Quang Huy, Minh N. H. Nguyen, Choong Seon Hong
Abstract: Many studies integrate federated learning (FL) with self-supervised learning (SSL) to take advantage of raw data distributed across edge devices. However, edge devices often struggle with high computation and communication costs imposed by SSL and FL algorithms. To tackle this hindrance, we propose LW-FedSSL, a layer-wise federated self-supervised learning approach that allows edge devices to incrementally train a single layer of the model at a time. We introduce server-side calibration and representation alignment mechanisms to ensure LW-FedSSL delivers performance on par with conventional federated self-supervised learning (FedSSL) while significantly lowering resource demands. In a pure layer-wise training scheme, training one layer at a time may limit effective interaction between different layers of the model. The server-side calibration mechanism takes advantage of the resource-rich FL server to ensure smooth collaboration between different layers of the global model. During local training, the representation alignment mechanism encourages closeness between representations of local models and those of the global model, thereby preserving the layer cohesion established by server-side calibration. With the proposed mechanisms, LW-FedSSL achieves a $3.3 \times$ reduction in memory usage, $2.1 \times$ fewer computational operations (FLOPs), and a $3.2 \times$ lower communication cost while maintaining the same level of performance as its end-to-end training counterpart. Additionally, we explore a progressive training strategy called Prog-FedSSL, which matches end-to-end training in memory requirements but offers a $1.8 \times$ reduction in FLOPs and communication costs. Although Prog-FedSSL is not as resource-efficient as LW-FedSSL, its performance improvements make it a suitable candidate for FL environments with more lenient resource constraints.
Authors: Andrea Apicella, Francesco Isgr\`o, Roberto Prevete
Abstract: Machine Learning (ML) has revolutionized various domains, offering predictive capabilities in several areas. However, with the increasing accessibility of ML tools, many practitioners, lacking deep ML expertise, adopt a "push the button" approach, utilizing user-friendly interfaces without a thorough understanding of underlying algorithms. While this approach provides convenience, it raises concerns about the reliability of outcomes, leading to challenges such as incorrect performance evaluation. This paper addresses a critical issue in ML, known as data leakage, where unintended information contaminates the training data, impacting model performance evaluation. Users, due to a lack of understanding, may inadvertently overlook crucial steps, leading to optimistic performance estimates that may not hold in real-world scenarios. The discrepancy between evaluated and actual performance on new data is a significant concern. In particular, this paper categorizes data leakage in ML, discussing how certain conditions can propagate through the ML workflow. Furthermore, it explores the connection between data leakage and the specific task being addressed, investigates its occurrence in Transfer Learning, and compares standard inductive ML with transductive ML frameworks. The conclusion summarizes key findings, emphasizing the importance of addressing data leakage for robust and reliable ML applications.
Authors: Kun Fang, Qinghua Tao, Kexin Lv, Mingzhen He, Xiaolin Huang, Jie Yang
Abstract: Out-of-Distribution (OoD) detection is vital for the reliability of Deep Neural Networks (DNNs). Existing works have shown the insufficiency of Principal Component Analysis (PCA) straightforwardly applied on the features of DNNs in detecting OoD data from In-Distribution (InD) data. The failure of PCA suggests that the network features residing in OoD and InD are not well separated by simply proceeding in a linear subspace, which instead can be resolved through proper non-linear mappings. In this work, we leverage the framework of Kernel PCA (KPCA) for OoD detection, and seek suitable non-linear kernels that advocate the separability between InD and OoD data in the subspace spanned by the principal components. Besides, explicit feature mappings induced from the devoted task-specific kernels are adopted so that the KPCA reconstruction error for new test samples can be efficiently obtained with large-scale data. Extensive theoretical and empirical results on multiple OoD data sets and network structures verify the superiority of our KPCA detector in efficiency and efficacy with state-of-the-art detection performance.
Authors: Amit Attia, Tomer Koren
Abstract: We study the problem of parameter-free stochastic optimization, inquiring whether, and under what conditions, do fully parameter-free methods exist: these are methods that achieve convergence rates competitive with optimally tuned methods, without requiring significant knowledge of the true problem parameters. Existing parameter-free methods can only be considered ``partially'' parameter-free, as they require some non-trivial knowledge of the true problem parameters, such as a bound on the stochastic gradient norms, a bound on the distance to a minimizer, etc. In the non-convex setting, we demonstrate that a simple hyperparameter search technique results in a fully parameter-free method that outperforms more sophisticated state-of-the-art algorithms. We also provide a similar result in the convex setting with access to noisy function values under mild noise assumptions. Finally, assuming only access to stochastic gradients, we establish a lower bound that renders fully parameter-free stochastic convex optimization infeasible, and provide a method which is (partially) parameter-free up to the limit indicated by our lower bound.
Authors: Anian Ruoss, Gr\'egoire Del\'etang, Sourabh Medapati, Jordi Grau-Moya, Li Kevin Wenliang, Elliot Catt, John Reid, Cannada A. Lewis, Joel Veness, Tim Genewein
Abstract: This paper uses chess, a landmark planning problem in AI, to assess transformers' performance on a planning task where memorization is futile $\unicode{x2013}$ even at a large scale. To this end, we release ChessBench, a large-scale benchmark dataset of 10 million chess games with legal move and value annotations (15 billion data points) provided by Stockfish 16, the state-of-the-art chess engine. We train transformers with up to 270 million parameters on ChessBench via supervised learning and perform extensive ablations to assess the impact of dataset size, model size, architecture type, and different prediction targets (state-values, action-values, and behavioral cloning). Our largest models learn to predict action-values for novel boards quite accurately, implying highly non-trivial generalization. Despite performing no explicit search, our resulting chess policy solves challenging chess puzzles and achieves a surprisingly strong Lichess blitz Elo of 2895 against humans (grandmaster level). We also compare to Leela Chess Zero and AlphaZero (trained without supervision via self-play) with and without search. We show that, although a remarkably good approximation of Stockfish's search-based algorithm can be distilled into large-scale transformers via supervised learning, perfect distillation is still beyond reach, thus making ChessBench well-suited for future research.
Authors: Chengxi Zeng, Tilo Burghardt, Alberto M Gambaruto
Abstract: In this paper, the training dynamics of PINNs with a feature mapping layer via the limiting Conjugate Kernel and Neural Tangent Kernel is investigated, shedding light on the convergence of PINNs; Although the commonly used Fourier-based feature mapping has achieved great success, we show its inadequacy in some physics scenarios. Via these two scopes, we propose conditionally positive definite Radial Basis Function as a better alternative. Lastly, we explore the feature mapping numerically in wide neural networks. Our empirical results reveal the efficacy of our method in diverse forward and inverse problem sets. Composing feature functions is found to be a practical way to address the expressivity and generalisability trade-off, viz., tuning the bandwidth of the kernels and the surjectivity of the feature mapping function. This simple technique can be implemented for coordinate inputs and benefits the broader PINNs research.
Authors: Junhan Kim, Chungman Lee, Eulrang Cho, Kyungphil Park, Ho-young Kim, Joonyoung Kim, Yongkweon Jeon
Abstract: With the increasing complexity of generative AI models, post-training quantization (PTQ) has emerged as a promising solution for deploying hyper-scale models on edge devices such as mobile and TVs. Existing PTQ schemes, however, consume considerable time and resources, which could be a bottleneck in real situations where frequent model updates and multiple hyperparameter tunings are required. As a cost-effective alternative, learning-free PTQ schemes have been proposed. However, the performance is somewhat limited because they cannot consider the inter-layer dependency within the attention module, which is a significant feature of Transformers. In this paper, we thus propose a novel PTQ algorithm that balances accuracy and efficiency. The key idea of the proposed algorithm called aespa is to perform quantization layer-wise for efficiency while targeting attention-wise reconstruction to consider the cross-layer dependency. Through extensive experiments on various language models and complexity analysis, we demonstrate that aespa is accurate and efficient in quantizing Transformer models.
Authors: Benjamin Feuer, Robin Tibor Schirrmeister, Valeriia Cherepanova, Chinmay Hegde, Frank Hutter, Micah Goldblum, Niv Cohen, Colin White
Abstract: While tabular classification has traditionally relied on from-scratch training, a recent breakthrough called prior-data fitted networks (PFNs) challenges this approach. Similar to large language models, PFNs make use of pretraining and in-context learning to achieve strong performance on new tasks in a single forward pass. However, current PFNs have limitations that prohibit their widespread adoption. Notably, TabPFN achieves very strong performance on small tabular datasets but is not designed to make predictions for datasets of size larger than 1000. In this work, we overcome these limitations and substantially improve the performance of PFNs via context optimization. We introduce TuneTables, a parameter-efficient fine-tuning strategy for PFNs that compresses large datasets into a smaller learned context. We conduct extensive experiments on 19 algorithms over 98 datasets and find that TuneTables achieves the best performance on average, outperforming boosted trees such as CatBoost, while optimizing fewer than 5% of TabPFN's parameters. Furthermore, we show that TuneTables can be used as an interpretability tool and can even be used to mitigate biases by optimizing a fairness objective. We open-source our code and raw results at https://github.com/penfever/TuneTables.
Authors: Ehsan Imani, Kai Luedemann, Sam Scholnick-Hughes, Esraa Elelimy, Martha White
Abstract: It is becoming increasingly common in regression to train neural networks that model the entire distribution even if only the mean is required for prediction. This additional modeling often comes with performance gain and the reasons behind the improvement are not fully known. This paper investigates a recent approach to regression, the Histogram Loss, which involves learning the conditional distribution of the target variable by minimizing the cross-entropy between a target distribution and a flexible histogram prediction. We design theoretical and empirical analyses to determine why and when this performance gain appears, and how different components of the loss contribute to it. Our results suggest that the benefits of learning distributions in this setup come from improvements in optimization rather than modelling extra information. We then demonstrate the viability of the Histogram Loss in common deep learning applications without a need for costly hyperparameter tuning.
Authors: Daniil Dmitriev, Krist\'of Szab\'o, Amartya Sanyal
Abstract: In this paper, we provide lower bounds for Differentially Private (DP) Online Learning algorithms. Our result shows that, for a broad class of $(\varepsilon,\delta)$-DP online algorithms, for number of rounds $T$ such that $\log T\leq O(1 / \delta)$, the expected number of mistakes incurred by the algorithm grows as $\Omega(\log \frac{T}{\delta})$. This matches the upper bound obtained by Golowich and Livni (2021) and is in contrast to non-private online learning where the number of mistakes is independent of $T$. To the best of our knowledge, our work is the first result towards settling lower bounds for DP-Online learning and partially addresses the open question in Sanyal and Ramponi (2022).
Authors: Yushun Zhang, Congliang Chen, Tian Ding, Ziniu Li, Ruoyu Sun, Zhi-Quan Luo
Abstract: SGD performs worse than Adam by a significant margin on Transformers, but the reason remains unclear. In this work, we provide an explanation through the lens of Hessian: (i) Transformers are "heterogeneous": the Hessian spectrum across parameter blocks vary dramatically, a phenomenon we call "block heterogeneity"; (ii) Heterogeneity hampers SGD: SGD performs worse than Adam on problems with block heterogeneity. To validate (i) and (ii), we check various Transformers, CNNs, MLPs, and quadratic problems, and find that SGD can perform on par with Adam on problems without block heterogeneity, but performs worse than Adam when the heterogeneity exists. Our initial theoretical analysis indicates that SGD performs worse because it applies one single learning rate to all blocks, which cannot handle the heterogeneity among blocks. This limitation could be ameliorated if we use coordinate-wise learning rates, as designed in Adam.
Authors: Lingkai Kong, Yuanqi Du, Wenhao Mu, Kirill Neklyudov, Valentin De Bortoli, Dongxia Wu, Haorui Wang, Aaron Ferber, Yi-An Ma, Carla P. Gomes, Chao Zhang
Abstract: Addressing real-world optimization problems becomes particularly challenging when analytic objective functions or constraints are unavailable. While numerous studies have addressed the issue of unknown objectives, limited research has focused on scenarios where feasibility constraints are not given explicitly. Overlooking these constraints can lead to spurious solutions that are unrealistic in practice. To deal with such unknown constraints, we propose to perform optimization within the data manifold using diffusion models. To constrain the optimization process to the data manifold, we reformulate the original optimization problem as a sampling problem from the product of the Boltzmann distribution defined by the objective function and the data distribution learned by the diffusion model. Depending on the differentiability of the objective function, we propose two different sampling methods. For differentiable objectives, we propose a two-stage framework that begins with a guided diffusion process for warm-up, followed by a Langevin dynamics stage for further correction. For non-differentiable objectives, we propose an iterative importance sampling strategy using the diffusion model as the proposal distribution. Comprehensive experiments on a synthetic dataset, six real-world black-box optimization datasets, and a multi-objective molecule optimization dataset show that our method achieves better or comparable performance with previous state-of-the-art baselines.
Authors: Xinyu Yuan, Yan Qiao
Abstract: Denoising diffusion probabilistic models (DDPMs) are becoming the leading paradigm for generative models. It has recently shown breakthroughs in audio synthesis, time series imputation and forecasting. In this paper, we propose Diffusion-TS, a novel diffusion-based framework that generates multivariate time series samples of high quality by using an encoder-decoder transformer with disentangled temporal representations, in which the decomposition technique guides Diffusion-TS to capture the semantic meaning of time series while transformers mine detailed sequential information from the noisy model input. Different from existing diffusion-based approaches, we train the model to directly reconstruct the sample instead of the noise in each diffusion step, combining a Fourier-based loss term. Diffusion-TS is expected to generate time series satisfying both interpretablity and realness. In addition, it is shown that the proposed Diffusion-TS can be easily extended to conditional generation tasks, such as forecasting and imputation, without any model changes. This also motivates us to further explore the performance of Diffusion-TS under irregular settings. Finally, through qualitative and quantitative experiments, results show that Diffusion-TS achieves the state-of-the-art results on various realistic analyses of time series.
Authors: DengYu Shi
Abstract: With the growing demand for energy and increased environmental awareness, Non-Intrusive Load Monitoring (NILM) has become an essential tool in smart grid and energy management. By analyzing total power load data, NILM infers the energy usage of individual appliances without the need for separate sensors, enabling real-time monitoring from a few locations. This approach helps users understand consumption patterns, enhance energy efficiency, and detect anomalies for effective energy management. However, NILM datasets often suffer from issues such as sensor failures and data loss, compromising data integrity, thereby impacting subsequent analysis and applications. Traditional imputation methods, such as linear interpolation and matrix factorization, struggle with nonlinear relationships and are sensitive to sparse data, resulting in information loss. To address these challenges, this paper proposes a Proportional-Integral-Derivative (PID) Controlled Non-Negative Latent Factorization of Tensor (PNLF) model, which dynamically adjusts parameter gradients to improve convergence, stability, and accuracy. Experimental results show that the PNLF model significantly outperforms state-of-the-art tensor completion models in both accuracy and efficiency. By addressing data loss issues, this study enhances load disaggregation precision and optimizes energy management, providing reliable data support for smart grid applications and policy formulation.
Authors: Aozhong Zhang, Zi Yang, Naigang Wang, Yingyong Qi, Jack Xin, Xin Li, Penghang Yin
Abstract: Post-training quantization (PTQ) has emerged as a practical approach to compress large neural networks, making them highly efficient for deployment. However, effectively reducing these models to their low-bit counterparts without compromising the original accuracy remains a key challenge. In this paper, we propose an innovative PTQ algorithm termed COMQ, which sequentially conducts coordinate-wise minimization of the layer-wise reconstruction errors. We consider the widely used integer quantization, where every quantized weight can be decomposed into a shared floating-point scalar and an integer bit-code. Within a fixed layer, COMQ treats all the scaling factor(s) and bit-codes as the variables of the reconstruction error. Every iteration improves this error along a single coordinate while keeping all other variables constant. COMQ is easy to use and requires no hyper-parameter tuning. It instead involves only dot products and rounding operations. We update these variables in a carefully designed greedy order, significantly enhancing the accuracy. COMQ achieves remarkable results in quantizing 4-bit Vision Transformers, with a negligible loss of less than 1% in Top-1 accuracy. In 4-bit INT quantization of convolutional neural networks, COMQ maintains near-lossless accuracy with a minimal drop of merely 0.3% in Top-1 accuracy.
Authors: Minhyuk Seo, Seongwon Cho, Minjae Lee, Diganta Misra, Hyeonbeom Choi, Seon Joo Kim, Jonghyun Choi
Abstract: Requiring extensive human supervision is often impractical for continual learning due to its cost, leading to the emergence of 'name-only continual learning' that only provides the name of new concepts (e.g., classes) without providing supervised samples. To address the task, recent approach uses web-scraped data but results in issues such as data imbalance, copyright, and privacy concerns. To overcome the limitations of both human supervision and webly supervision, we propose Generative name only Continual Learning (GenCL) using generative models for the name only continual learning. But na\"ive application of generative models results in limited diversity of generated data. So, we specifically propose a diverse prompt generation method, HIerarchical Recurrent Prompt Generation (HIRPG) as well as COmplexity-NAvigating eNsembler (CONAN) that selects samples with minimal overlap from multiple generative models. We empirically validate that the proposed GenCL outperforms prior arts, even a model trained with fully supervised data, in various tasks including image recognition and multi-modal visual reasoning. Data generated by GenCL is available at https://anonymous.4open.science/r/name-only-continual-E079.
URLs: https://anonymous.4open.science/r/name-only-continual-E079.
Authors: Gregory Dexter, Christos Boutsikas, Linkai Ma, Ilse C. F. Ipsen, Petros Drineas
Abstract: Motivated by the popularity of stochastic rounding in the context of machine learning and the training of large-scale deep neural network models, we consider stochastic nearness rounding of real matrices $\mathbf{A}$ with many more rows than columns. We provide novel theoretical evidence, supported by extensive experimental evaluation that, with high probability, the smallest singular value of a stochastically rounded matrix is well bounded away from zero -- regardless of how close $\mathbf{A}$ is to being rank deficient and even if $\mathbf{A}$ is rank-deficient. In other words, stochastic rounding \textit{implicitly regularizes} tall and skinny matrices $\mathbf{A}$ so that the rounded version has full column rank. Our proofs leverage powerful results in random matrix theory, and the idea that stochastic rounding errors do not concentrate in low-dimensional column spaces.
Authors: Artem Vysogorets, Kartik Ahuja, Julia Kempe
Abstract: In the era of exceptionally data-hungry models, careful selection of the training data is essential to mitigate the extensive costs of deep learning. Data pruning offers a solution by removing redundant or uninformative samples from the dataset, which yields faster convergence and improved neural scaling laws. However, little is known about its impact on classification bias of the trained models. We conduct the first systematic study of this effect and reveal that existing data pruning algorithms can produce highly biased classifiers. We present theoretical analysis of the classification risk in a mixture of Gaussians to argue that choosing appropriate class pruning ratios, coupled with random pruning within classes has potential to improve worst-class performance. We thus propose DRoP, a distributionally robust approach to pruning and empirically demonstrate its performance on standard computer vision benchmarks. In sharp contrast to existing algorithms, our proposed method continues improving distributional robustness at a tolerable drop of average performance as we prune more from the datasets.
Authors: Akifumi Wachi, Thien Q. Tran, Rei Sato, Takumi Tanabe, Youhei Akimoto
Abstract: Safety and trustworthiness are indispensable requirements for real-world applications of AI systems using large language models (LLMs). This paper formulates human value alignment as an optimization problem of the language model policy to maximize reward under a safety constraint, and then proposes an algorithm, Stepwise Alignment for Constrained Policy Optimization (SACPO). One key idea behind SACPO, supported by theory, is that the optimal policy incorporating reward and safety can be directly obtained from a reward-aligned policy. Building on this key idea, SACPO aligns LLMs step-wise with each metric while leveraging simple yet powerful alignment algorithms such as direct preference optimization (DPO). SACPO offers several advantages, including simplicity, stability, computational efficiency, and flexibility of algorithms and datasets. Under mild assumptions, our theoretical analysis provides the upper bounds on optimality and safety constraint violation. Our experimental results show that SACPO can fine-tune Alpaca-7B better than the state-of-the-art method in terms of both helpfulness and harmlessness.
Authors: Natalie S. Frank
Abstract: Minimizing an adversarial surrogate risk is a common technique for learning robust classifiers. Prior work showed that convex surrogate losses are not statistically consistent in the adversarial context -- or in other words, a minimizing sequence of the adversarial surrogate risk will not necessarily minimize the adversarial classification error. We connect the consistency of adversarial surrogate losses to properties of minimizers to the adversarial classification risk, known as adversarial Bayes classifiers. Specifically, under reasonable distributional assumptions, a convex surrogate loss is statistically consistent for adversarial learning iff the adversarial Bayes classifier satisfies a certain notion of uniqueness.
Authors: David J. Schodt
Abstract: Bayesian Neural Networks (BNNs) extend traditional neural networks to provide uncertainties associated with their outputs. On the forward pass through a BNN, predictions (and their uncertainties) are made either by Monte Carlo sampling network weights from the learned posterior or by analytically propagating statistical moments through the network. Though flexible, Monte Carlo sampling is computationally expensive and can be infeasible or impractical under resource constraints or for large networks. While moment propagation can ameliorate the computational costs of BNN inference, it can be difficult or impossible for networks with arbitrary nonlinearities, thereby restricting the possible set of network layers permitted with such a scheme. In this work, we demonstrate a simple yet effective approach for propagating statistical moments through arbitrary nonlinearities with only 3 deterministic samples, enabling few-sample variational inference of BNNs without restricting the set of network layers used. Furthermore, we leverage this approach to demonstrate a novel nonlinear activation function that we use to inject physics-informed prior information into output nodes of a BNN.
Authors: Zine el abidine Kherroubi
Abstract: Current autonomous driving systems heavily rely on V2X communication data to enhance situational awareness and the cooperation between vehicles. However, a major challenge when using V2X data is that it may not be available periodically because of unpredictable delays and data loss during wireless transmission between road stations and the receiver vehicle. This issue should be considered when designing control strategies for connected and autonomous vehicles. Therefore, this paper proposes a novel 'Blind Actor-Critic' algorithm that guarantees robust driving performance in V2X environment with delayed and/or lost data. The novel algorithm incorporates three key mechanisms: a virtual fixed sampling period, a combination of Temporal-Difference and Monte Carlo learning, and a numerical approximation of immediate reward values. To address the temporal aperiodicity problem of V2X data, we first illustrate this challenge. Then, we provide a detailed explanation of the Blind Actor-Critic algorithm where we highlight the proposed components to compensate for the temporal aperiodicity problem of V2X data. We evaluate the performance of our algorithm in a simulation environment and compare it to benchmark approaches. The results demonstrate that training metrics are improved compared to conventional actor-critic algorithms. Additionally, testing results show that our approach provides robust control, even under low V2X network reliability levels.
Authors: Tongze Wang, Xiaohui Xie, Wenduo Wang, Chuyi Wang, Youjian Zhao, Yong Cui
Abstract: Network traffic classification is a crucial research area aiming to enhance service quality, streamline network management, and bolster cybersecurity. To address the growing complexity of transmission encryption techniques, various machine learning and deep learning methods have been proposed. However, existing approaches face two main challenges. Firstly, they struggle with model inefficiency due to the quadratic complexity of the widely used Transformer architecture. Secondly, they suffer from inadequate traffic representation because of discarding important byte information while retaining unwanted biases. To address these challenges, we propose NetMamba, an efficient linear-time state space model equipped with a comprehensive traffic representation scheme. We adopt a specially selected and improved unidirectional Mamba architecture for the networking field, instead of the Transformer, to address efficiency issues. In addition, we design a traffic representation scheme to extract valid information from massive traffic data while removing biased information. Evaluation experiments on six public datasets encompassing three main classification tasks showcase NetMamba's superior classification performance compared to state-of-the-art baselines. It achieves an accuracy rate of nearly 99% (some over 99%) in all tasks. Additionally, NetMamba demonstrates excellent efficiency, improving inference speed by up to 60 times while maintaining comparably low memory usage. Furthermore, NetMamba exhibits superior few-shot learning abilities, achieving better classification performance with fewer labeled data. To the best of our knowledge, NetMamba is the first model to tailor the Mamba architecture for networking.
Authors: Yijin Ni, Xiaoming Huo
Abstract: Maximum Mean Discrepancy (MMD) is a probability metric that has found numerous applications in machine learning. In this work, we focus on its application in generative models, including the minimum MMD estimator, Generative Moment Matching Network (GMMN), and Generative Adversarial Network (GAN). In these cases, MMD is part of an objective function in a minimization or min-max optimization problem. Even if its empirical performance is competitive, the consistency and convergence rate analysis of the corresponding MMD-based estimators has yet to be carried out. We propose a uniform concentration inequality for a class of Maximum Mean Discrepancy (MMD)-based estimators, that is, a maximum deviation bound of empirical MMD values over a collection of generated distributions and adversarially learned kernels. Here, our inequality serves as an efficient tool in the theoretical analysis for MMD-based generative models. As elaborating examples, we applied our main result to provide the generalization error bounds for the MMD-based estimators in the context of the minimum MMD estimator and MMD GAN.
Authors: Haixu Wu, Huakun Luo, Yuezhou Ma, Jianmin Wang, Mingsheng Long
Abstract: Physics-informed neural networks (PINNs) have been widely applied to solve partial differential equations (PDEs) by enforcing outputs and gradients of deep models to satisfy target equations. Due to the limitation of numerical computation, PINNs are conventionally optimized on finite selected points. However, since PDEs are usually defined on continuous domains, solely optimizing models on scattered points may be insufficient to obtain an accurate solution for the whole domain. To mitigate this inherent deficiency of the default scatter-point optimization, this paper proposes and theoretically studies a new training paradigm as region optimization. Concretely, we propose to extend the optimization process of PINNs from isolated points to their continuous neighborhood regions, which can theoretically decrease the generalization error, especially for hidden high-order constraints of PDEs. A practical training algorithm, Region Optimized PINN (RoPINN), is seamlessly derived from this new paradigm, which is implemented by a straightforward but effective Monte Carlo sampling method. By calibrating the sampling process into trust regions, RoPINN finely balances optimization and generalization error. Experimentally, RoPINN consistently boosts the performance of diverse PINNs on a wide range of PDEs without extra backpropagation or gradient calculation. Code is available at this repository: https://github.com/thuml/RoPINN.
Authors: Peter S\'uken\'ik, Marco Mondelli, Christoph Lampert
Abstract: Deep neural networks (DNNs) exhibit a surprising structure in their final layer known as neural collapse (NC), and a growing body of works has currently investigated the propagation of neural collapse to earlier layers of DNNs -- a phenomenon called deep neural collapse (DNC). However, existing theoretical results are restricted to special cases: linear models, only two layers or binary classification. In contrast, we focus on non-linear models of arbitrary depth in multi-class classification and reveal a surprising qualitative shift. As soon as we go beyond two layers or two classes, DNC stops being optimal for the deep unconstrained features model (DUFM) -- the standard theoretical framework for the analysis of collapse. The main culprit is a low-rank bias of multi-layer regularization schemes: this bias leads to optimal solutions of even lower rank than the neural collapse. We support our theoretical findings with experiments on both DUFM and real data, which show the emergence of the low-rank structure in the solution found by gradient descent.
Authors: Sujai Hiremath, Jacqueline R. M. A. Maasch, Mengxiao Gao, Promit Ghosal, Kyra Gan
Abstract: Learning the unique directed acyclic graph corresponding to an unknown causal model is a challenging task. Methods based on functional causal models can identify a unique graph, but either suffer from the curse of dimensionality or impose strong parametric assumptions. To address these challenges, we propose a novel hybrid approach for global causal discovery in observational data that leverages local causal substructures. We first present a topological sorting algorithm that leverages ancestral relationships in linear structural equation models to establish a compact top-down hierarchical ordering, encoding more causal information than linear orderings produced by existing methods. We demonstrate that this approach generalizes to nonlinear settings with arbitrary noise. We then introduce a nonparametric constraint-based algorithm that prunes spurious edges by searching for local conditioning sets, achieving greater accuracy than current methods. We provide theoretical guarantees for correctness and worst-case polynomial time complexities, with empirical validation on synthetic data.
Authors: Th\'eo Vincent, Fabian Wahren, Jan Peters, Boris Belousov, Carlo D'Eramo
Abstract: Deep Reinforcement Learning (RL) is well known for being highly sensitive to hyperparameters, requiring practitioners substantial efforts to optimize them for the problem at hand. This also limits the applicability of RL in real-world scenarios. In recent years, the field of automated Reinforcement Learning (AutoRL) has grown in popularity by trying to address this issue. However, these approaches typically hinge on additional samples to select well-performing hyperparameters, hindering sample-efficiency and practicality. Furthermore, most AutoRL methods are heavily based on already existing AutoML methods, which were originally developed neglecting the additional challenges inherent to RL due to its non-stationarities. In this work, we propose a new approach for AutoRL, called Adaptive $Q$-Network (AdaQN), that is tailored to RL to take into account the non-stationarity of the optimization procedure without requiring additional samples. AdaQN learns several $Q$-functions, each one trained with different hyperparameters, which are updated online using the $Q$-function with the smallest approximation error as a shared target. Our selection scheme simultaneously handles different hyperparameters while coping with the non-stationarity induced by the RL optimization procedure and being orthogonal to any critic-based RL algorithm. We demonstrate that AdaQN is theoretically sound and empirically validate it in MuJoCo control problems and Atari $2600$ games, showing benefits in sample-efficiency, overall performance, robustness to stochasticity and training stability.
Authors: Yuntong Hu, Zhihan Lei, Zheng Zhang, Bo Pan, Chen Ling, Liang Zhao
Abstract: Naive Retrieval-Augmented Generation (RAG) focuses on individual documents during retrieval and, as a result, falls short in handling networked documents which are very popular in many applications such as citation graphs, social media, and knowledge graphs. To overcome this limitation, we introduce Graph Retrieval-Augmented Generation (GRAG), which tackles the fundamental challenges in retrieving textual subgraphs and integrating the joint textual and topological information into Large Language Models (LLMs) to enhance its generation. To enable efficient textual subgraph retrieval, we propose a novel divide-and-conquer strategy that retrieves the optimal subgraph structure in linear time. To achieve graph context-aware generation, incorporate textual graphs into LLMs through two complementary views-the text view and the graph view-enabling LLMs to more effectively comprehend and utilize the graph context. Extensive experiments on graph reasoning benchmarks demonstrate that in scenarios requiring multi-hop reasoning on textual graphs, our GRAG approach significantly outperforms current state-of-the-art RAG methods.
Authors: Ziao Yang, Han Yue, Jian Chen, Hongfu Liu
Abstract: Influence functions serve as crucial tools for assessing sample influence in model interpretation, subset training set selection, noisy label detection, and more. By employing the first-order Taylor extension, influence functions can estimate sample influence without the need for expensive model retraining. However, applying influence functions directly to deep models presents challenges, primarily due to the non-convex nature of the loss function and the large size of model parameters. This difficulty not only makes computing the inverse of the Hessian matrix costly but also renders it non-existent in some cases. Various approaches, including matrix decomposition, have been explored to expedite and approximate the inversion of the Hessian matrix, with the aim of making influence functions applicable to deep models. In this paper, we revisit a specific, albeit naive, yet effective approximation method known as TracIn. This method substitutes the inverse of the Hessian matrix with an identity matrix. We provide deeper insights into why this simple approximation method performs well. Furthermore, we extend its applications beyond measuring model utility to include considerations of fairness and robustness. Finally, we enhance TracIn through an ensemble strategy. To validate its effectiveness, we conduct experiments on synthetic data and extensive evaluations on noisy label detection, sample selection for large language model fine-tuning, and defense against adversarial attacks.
Authors: Dongjae Jeon, Wonje Jeung, Taeheon Kim, Albert No, Jonghyun Choi
Abstract: Machine unlearning (MU) aims to remove the influence of specific data from trained models, addressing privacy concerns and ensuring compliance with regulations such as the "right to be forgotten." Evaluating strong unlearning, where the unlearned model is indistinguishable from one retrained without the forgetting data, remains a significant challenge in deep neural networks (DNNs). Common black-box metrics, such as variants of membership inference attacks and accuracy comparisons, primarily assess model outputs but often fail to capture residual information in intermediate layers. To bridge this gap, we introduce the Information Difference Index (IDI), a novel white-box metric inspired by information theory. IDI quantifies retained information in intermediate features by measuring mutual information between those features and the labels to be forgotten, offering a more comprehensive assessment of unlearning efficacy. Our experiments demonstrate that IDI effectively measures the degree of unlearning across various datasets and architectures, providing a reliable tool for evaluating strong unlearning in DNNs.
Authors: Yang Zhang, Yawei Li, Xinpeng Wang, Qianli Shen, Barbara Plank, Bernd Bischl, Mina Rezaei, Kenji Kawaguchi
Abstract: Overparametrized transformer networks are the state-of-the-art architecture for Large Language Models (LLMs). However, such models contain billions of parameters making large compute a necessity, while raising environmental concerns. To address these issues, we propose FinerCut, a new form of fine-grained layer pruning, which in contrast to prior work at the transformer block level, considers all self-attention and feed-forward network (FFN) layers within blocks as individual pruning candidates. FinerCut prunes layers whose removal causes minimal alternation to the model's output -- contributing to a new, lean, interpretable, and task-agnostic pruning method. Tested across 9 benchmarks, our approach retains 90% performance of Llama3-8B with 25% layers removed, and 95% performance of Llama3-70B with 30% layers removed, all without fine-tuning or post-pruning reconstruction. Strikingly, we observe intriguing results with FinerCut: 42% (34 out of 80) of the self-attention layers in Llama3-70B can be removed while preserving 99% of its performance -- without additional fine-tuning after removal. Moreover, FinerCut provides a tool to inspect the types and locations of pruned layers, allowing to observe interesting pruning behaviors. For instance, we observe a preference for pruning self-attention layers, often at deeper consecutive decoder layers. We hope our insights inspire future efficient LLM architecture designs.
Authors: Hangzhi Guo, Firdaus Ahmed Choudhury, Tinghua Chen, Amulya Yadav
Abstract: Counterfactual (CF) explanations for ML model predictions provide actionable recourse recommendations to individuals adversely impacted by predicted outcomes. However, despite being preferred by end-users, CF explanations have been shown to pose significant security risks in real-world applications; in particular, malicious adversaries can exploit CF explanations to perform query-efficient model extraction attacks on the underlying proprietary ML model. To address this security challenge, we propose CFMark, a novel model-agnostic watermarking framework for detecting unauthorized model extraction attacks relying on CF explanations. CFMark involves a novel bi-level optimization problem to embed an indistinguishable watermark into the generated CF explanation such that any future model extraction attacks using these watermarked CF explanations can be detected using a null hypothesis significance testing (NHST) scheme. At the same time, the embedded watermark does not compromise the quality of the CF explanations. We evaluate CFMark across diverse real-world datasets, CF explanation methods, and model extraction techniques. Our empirical results demonstrate CFMark's effectiveness, achieving an F-1 score of ~0.89 in identifying unauthorized model extraction attacks using watermarked CF explanations. Importantly, this watermarking incurs only a negligible degradation in the quality of generated CF explanations (i.e., ~1.3% degradation in validity and ~1.6% in proximity). Our work establishes a critical foundation for the secure deployment of CF explanations in real-world applications.
Authors: Ethan Rathbun, Christopher Amato, Alina Oprea
Abstract: Reinforcement learning (RL) is an actively growing field that is seeing increased usage in real-world, safety-critical applications -- making it paramount to ensure the robustness of RL algorithms against adversarial attacks. In this work we explore a particularly stealthy form of training-time attacks against RL -- backdoor poisoning. Here the adversary intercepts the training of an RL agent with the goal of reliably inducing a particular action when the agent observes a pre-determined trigger at inference time. We uncover theoretical limitations of prior work by proving their inability to generalize across domains and MDPs. Motivated by this, we formulate a novel poisoning attack framework which interlinks the adversary's objectives with those of finding an optimal policy -- guaranteeing attack success in the limit. Using insights from our theoretical analysis we develop ``SleeperNets'' as a universal backdoor attack which exploits a newly proposed threat model and leverages dynamic reward poisoning techniques. We evaluate our attack in 6 environments spanning multiple domains and demonstrate significant improvements in attack success over existing methods, while preserving benign episodic return.
Authors: Kieran A. Murphy, Sam Dillavou, Dani S. Bassett
Abstract: Probabilistic representation spaces convey information about a dataset, and to understand the effects of factors such as training loss and network architecture, we seek to compare the information content of such spaces. However, most existing methods to compare representation spaces assume representations are points, and neglect the distributional nature of probabilistic representations. Here, instead of building upon point-based measures of comparison, we build upon classic methods from literature on hard clustering. We generalize two information-theoretic methods of comparing hard clustering assignments to be applicable to general probabilistic representation spaces. We then propose a practical method of estimation that is based on fingerprinting a representation space with a sample of the dataset and is applicable when the communicated information is only a handful of bits. With unsupervised disentanglement as a motivating problem, we find information fragments that are repeatedly contained in individual latent dimensions in VAE and InfoGAN ensembles. Then, by comparing the full latent spaces of models, we find highly consistent information content across datasets, methods, and hyperparameters, even though there is often a point during training with substantial variety across repeat runs. Finally, we leverage the differentiability of the proposed method and perform model fusion by synthesizing the information content of multiple weak learners, each incapable of representing the global structure of a dataset. Across the case studies, the direct comparison of information content provides a natural basis for understanding the processing of information.
Authors: Louis Serrano, Thomas X Wang, Etienne Le Naour, Jean-No\"el Vittaut, Patrick Gallinari
Abstract: We present AROMA (Attentive Reduced Order Model with Attention), a framework designed to enhance the modeling of partial differential equations (PDEs) using local neural fields. Our flexible encoder-decoder architecture can obtain smooth latent representations of spatial physical fields from a variety of data types, including irregular-grid inputs and point clouds. This versatility eliminates the need for patching and allows efficient processing of diverse geometries. The sequential nature of our latent representation can be interpreted spatially and permits the use of a conditional transformer for modeling the temporal dynamics of PDEs. By employing a diffusion-based formulation, we achieve greater stability and enable longer rollouts compared to conventional MSE training. AROMA's superior performance in simulating 1D and 2D equations underscores the efficacy of our approach in capturing complex dynamical behaviors.
Authors: Nadav Merlis
Abstract: We study reinforcement learning (RL) problems in which agents observe the reward or transition realizations at their current state before deciding which action to take. Such observations are available in many applications, including transactions, navigation and more. When the environment is known, previous work shows that this lookahead information can drastically increase the collected reward. However, outside of specific applications, existing approaches for interacting with unknown environments are not well-adapted to these observations. In this work, we close this gap and design provably-efficient learning algorithms able to incorporate lookahead information. To achieve this, we perform planning using the empirical distribution of the reward and transition observations, in contrast to vanilla approaches that only rely on estimated expectations. We prove that our algorithms achieve tight regret versus a baseline that also has access to lookahead information - linearly increasing the amount of collected reward compared to agents that cannot handle lookahead information.
Authors: Tim Z. Xiao, Robert Bamler, Bernhard Sch\"olkopf, Weiyang Liu
Abstract: Motivated by the progress made by large language models (LLMs), we introduce the framework of verbalized machine learning (VML). In contrast to conventional machine learning (ML) models that are typically optimized over a continuous parameter space, VML constrains the parameter space to be human-interpretable natural language. Such a constraint leads to a new perspective of function approximation, where an LLM with a text prompt can be viewed as a function parameterized by the text prompt. Guided by this perspective, we revisit classical ML problems, such as regression and classification, and find that these problems can be solved by an LLM-parameterized learner and optimizer. The major advantages of VML include (1) easy encoding of inductive bias: prior knowledge about the problem and hypothesis class can be encoded in natural language and fed into the LLM-parameterized learner; (2) automatic model class selection: the optimizer can automatically select a model class based on data and verbalized prior knowledge, and it can update the model class during training; and (3) interpretable learner updates: the LLM-parameterized optimizer can provide explanations for why an update is performed. We empirically verify the effectiveness of VML, and hope that VML can serve as a stepping stone to stronger interpretability.
Authors: Lu Li, Tianyu Zhang, Zhiqi Bu, Suyuchen Wang, Huan He, Jie Fu, Yonghui Wu, Jiang Bian, Yong Chen, Yoshua Bengio
Abstract: Model merging has emerged as an effective approach to combine multiple single-task models into a multitask model. This process typically involves computing a weighted average of the model parameters without any additional training. Existing model-merging methods focus on enhancing average task accuracy. However, interference and conflicts between the objectives of different tasks can lead to trade-offs during the merging process. In real-world applications, a set of solutions with various trade-offs can be more informative, helping practitioners make decisions based on diverse preferences. In this paper, we introduce a novel and low-compute algorithm, Model Merging with Amortized Pareto Front (MAP). MAP efficiently identifies a Pareto set of scaling coefficients for merging multiple models, reflecting the trade-offs involved. It amortizes the substantial computational cost of evaluations needed to estimate the Pareto front by using quadratic approximation surrogate models derived from a pre-selected set of scaling coefficients. Experimental results on vision and natural language processing tasks demonstrate that MAP can accurately identify the Pareto front, providing practitioners with flexible solutions to balance competing task objectives. We also introduce Bayesian MAP for scenarios with a relatively low number of tasks and Nested MAP for situations with a high number of tasks, further reducing the computational cost of evaluation.
Authors: Mert Albaba, Sammy Christen, Thomas Langarek, Christoph Gebhardt, Otmar Hilliges, Michael J. Black
Abstract: Reinforcement Learning has achieved significant success in generating complex behavior but often requires extensive reward function engineering. Adversarial variants of Imitation Learning and Inverse Reinforcement Learning offer an alternative by learning policies from expert demonstrations via a discriminator. However, these methods struggle in complex tasks where randomly sampling expert-like behaviors is challenging. This limitation stems from their reliance on policy-agnostic discriminators, which provide insufficient guidance for agent improvement, especially as task complexity increases and expert behavior becomes more distinct. We introduce RILe (Reinforced Imitation Learning environment), a novel trainer-student system that learns a dynamic reward function based on the student's performance and alignment with expert demonstrations. In RILe, the student learns an action policy while the trainer, using reinforcement learning, continuously updates itself via the discriminator's feedback to optimize the alignment between the student and the expert. The trainer optimizes for long-term cumulative rewards from the discriminator, enabling it to provide nuanced feedback that accounts for the complexity of the task and the student's current capabilities. This approach allows for greater exploration of agent actions by providing graduated feedback rather than binary expert/non-expert classifications. By reducing dependence on policy-agnostic discriminators, RILe enables better performance in complex settings where traditional methods falter, outperforming existing methods by 2x in complex simulated robot-locomotion tasks.
Authors: Yuan Gao, Zujing Liu, Weizhong Zhang, Bo Du, Gui-Song Xia
Abstract: In contrast to moderate-size neural network pruning, structural weight pruning on the Large-Language Models (LLMs) imposes a novel challenge on the efficiency of the pruning algorithms, due to the heavy computation/memory demands of the LLMs. Recent efficient LLM pruning methods typically operate at the post-training phase without the expensive weight finetuning, however, their pruning criteria often rely on heuristically hand-crafted metrics, potentially leading to suboptimal performance. We instead propose a novel optimization-based structural pruning that learns the pruning masks in a probabilistic space directly by optimizing the loss of the pruned model. To preserve the efficiency, our method eliminates the back-propagation through the LLM per se during the optimization, requiring only the forward pass of the LLM. We achieve this by learning an underlying Bernoulli distribution to sample binary pruning masks, where we decouple the Bernoulli parameters from the LLM loss, thus facilitating an efficient optimization via a policy gradient estimator without back-propagation. As a result, our method is able to 1) operate at structural granularities of channels, heads, and layers, 2) support global and heterogeneous pruning (i.e., our method automatically determines different redundancy for different layers), and 3) optionally initialize with a metric-based method (for our Bernoulli distributions). Extensive experiments on LLaMA, LLaMA-2, LLaMA-3, Vicuna, and Mistral using the C4 and WikiText2 datasets demonstrate that our method operates for 2.7 hours with around 35GB memory for the 13B models on a single A100 GPU, and our pruned models outperform the state-of-the-arts w.r.t. both perplexity and the majority of various zero-shot tasks. Codes will be released.
Authors: Hongyi Zhou, Denis Blessing, Ge Li, Onur Celik, Xiaogang Jia, Gerhard Neumann, Rudolf Lioutikov
Abstract: This work introduces Variational Diffusion Distillation (VDD), a novel method that distills denoising diffusion policies into Mixtures of Experts (MoE) through variational inference. Diffusion Models are the current state-of-the-art in generative modeling due to their exceptional ability to accurately learn and represent complex, multi-modal distributions. This ability allows Diffusion Models to replicate the inherent diversity in human behavior, making them the preferred models in behavior learning such as Learning from Human Demonstrations (LfD). However, diffusion models come with some drawbacks, including the intractability of likelihoods and long inference times due to their iterative sampling process. The inference times, in particular, pose a significant challenge to real-time applications such as robot control. In contrast, MoEs effectively address the aforementioned issues while retaining the ability to represent complex distributions but are notoriously difficult to train. VDD is the first method that distills pre-trained diffusion models into MoE models, and hence, combines the expressiveness of Diffusion Models with the benefits of Mixture Models. Specifically, VDD leverages a decompositional upper bound of the variational objective that allows the training of each expert separately, resulting in a robust optimization scheme for MoEs. VDD demonstrates across nine complex behavior learning tasks, that it is able to: i) accurately distill complex distributions learned by the diffusion model, ii) outperform existing state-of-the-art distillation methods, and iii) surpass conventional methods for training MoE.
Authors: Jiarui Feng, Hao Liu, Lecheng Kong, Mingfang Zhu, Yixin Chen, Muhan Zhang
Abstract: In this report, we present TAGLAS, an atlas of text-attributed graph (TAG) datasets and benchmarks. TAGs are graphs with node and edge features represented in text, which have recently gained wide applicability in training graph-language or graph foundation models. In TAGLAS, we collect and integrate more than 23 TAG datasets with domains ranging from citation graphs to molecule graphs and tasks from node classification to graph question-answering. Unlike previous graph datasets and benchmarks, all datasets in TAGLAS have a unified node and edge text feature format, which allows a graph model to be simultaneously trained and evaluated on multiple datasets from various domains. Further, we provide a standardized, efficient, and simplified way to load all datasets and tasks. We also provide useful utils like text-to-embedding conversion, and graph-to-text conversion, which can facilitate different evaluation scenarios. Finally, we also provide standard and easy-to-use evaluation utils. The project is open-sourced at https://github.com/JiaruiFeng/TAGLAS and is still under construction. Please expect more datasets/features in the future.
Authors: Yulong Yang, Bowen Feng, Keqin Wang, Naomi Leonard, Adji Bousso Dieng, Christine Allen-Blanchette
Abstract: From pedestrians to Kuramoto oscillators, interactions between agents govern how a multitude of dynamical systems evolve in space and time. Discovering how these agents relate to each other can improve our understanding of the often complex dynamics that underlie these systems. Recent works learn to categorize relationships between agents based on observations of their physical behavior. These approaches are limited in that the relationship categories are modelled as outcomes of categorical distribution, when in real world systems categories often intermingle and interact. In this work, we introduce a level of abstraction between the observable behavior of agents and the latent categories that determine their behavior. To do this, we learn a mapping from agent behavior to agent preferences for each latent category in a graph neural network. We integrate the physical proximity of agents and their preferences in a nonlinear opinion dynamics model which provides a mechanism to identify mutually exclusive latent categories, predict an agent's evolution in time, and control an agent's physical behavior. We demonstrate the utility of our model for learning interpretable categories, and its efficacy on long-horizon prediction across several benchmarks where we outperform existing methods.
Authors: Qiushuo Hou, Matteo Zecchin, Sangwoo Park, Yunlong Cai, Guanding Yu, Kaushik Chowdhury, Osvaldo Simeone
Abstract: In modern wireless network architectures, such as O-RAN, artificial intelligence (AI)-based applications are deployed at intelligent controllers to carry out functionalities like scheduling or power control. The AI "apps" are selected on the basis of contextual information such as network conditions, topology, traffic statistics, and design goals. The mapping between context and AI model parameters is ideally done in a zero-shot fashion via an automatic model selection (AMS) mapping that leverages only contextual information without requiring any current data. This paper introduces a general methodology for the online optimization of AMS mappings. Optimizing an AMS mapping is challenging, as it requires exposure to data collected from many different contexts. Therefore, if carried out online, this initial optimization phase would be extremely time consuming. A possible solution is to leverage a digital twin of the physical system to generate synthetic data from multiple simulated contexts. However, given that the simulator at the digital twin is imperfect, a direct use of simulated data for the optimization of the AMS mapping would yield poor performance when tested in the real system. This paper proposes a novel method for the online optimization of AMS mapping that corrects for the bias of the simulator by means of limited real data collected from the physical system. Experimental results for a graph neural network-based power control app demonstrate the significant advantages of the proposed approach.
Authors: Abdur Rashid, Parag Biswas, Angona Biswas, MD Abdullah Al Nasim, Kishor Datta Gupta, Roy George
Abstract: Artificial intelligence (AI) has become a crucial instrument for streamlining processes in various industries, including electrical power systems, as a result of recent digitalization. Algorithms for artificial intelligence are data-driven models that are based on statistical learning theory and are used as a tool to take use of the data that the power system and its users generate. Initially, we perform a thorough literature analysis of artificial intelligence (AI) applications related to renewable energy (RE). Next, we present a thorough analysis of renewable energy factories and assess their suitability, along with a list of the most widely used and appropriate AI algorithms. Nine AI-based strategies are identified here to assist Renewable Energy (RE) in contemporary power systems. This survey paper comprises an extensive review of the several AI techniques used for renewable energy as well as a methodical analysis of the literature for the study of various intelligent system application domains across different disciplines of renewable energy. This literature review identifies the performance and outcomes of nine different research methods by assessing them, and it aims to distill valuable insights into their strengths and limitations. This study also addressed three main topics: using AI technology for renewable power generation, utilizing AI for renewable energy forecasting, and optimizing energy systems. Additionally, it explored AI's superiority over conventional models in controllability, data handling, cyberattack prevention, smart grid implementation, robotics- AI's significance in shaping the future of the energy industry. Furthermore, this article outlines future directions in the integration of AI for renewable energy.
Authors: Jeremias Traub, Till J. Bungert, Carsten T. L\"uth, Michael Baumgartner, Klaus H. Maier-Hein, Lena Maier-Hein, Paul F Jaeger
Abstract: Selective Classification, wherein models can reject low-confidence predictions, promises reliable translation of machine-learning based classification systems to real-world scenarios such as clinical diagnostics. While current evaluation of these systems typically assumes fixed working points based on pre-defined rejection thresholds, methodological progress requires benchmarking the general performance of systems akin to the $\mathrm{AUROC}$ in standard classification. In this work, we define 5 requirements for multi-threshold metrics in selective classification regarding task alignment, interpretability, and flexibility, and show how current approaches fail to meet them. We propose the Area under the Generalized Risk Coverage curve ($\mathrm{AUGRC}$), which meets all requirements and can be directly interpreted as the average risk of undetected failures. We empirically demonstrate the relevance of $\mathrm{AUGRC}$ on a comprehensive benchmark spanning 6 data sets and 13 confidence scoring functions. We find that the proposed metric substantially changes metric rankings on 5 out of the 6 data sets.
Authors: Parham Mohammad Panahi, Andrew Patterson, Martha White, Adam White
Abstract: Experience replay, the reuse of past data to improve sample efficiency, is ubiquitous in reinforcement learning. Though a variety of smart sampling schemes have been introduced to improve performance, uniform sampling by far remains the most common approach. One exception is Prioritized Experience Replay (PER), where sampling is done proportionally to TD errors, inspired by the success of prioritized sweeping in dynamic programming. The original work on PER showed improvements in Atari, but follow-up results were mixed. In this paper, we investigate several variations on PER, to attempt to understand where and when PER may be useful. Our findings in prediction tasks reveal that while PER can improve value propagation in tabular settings, behavior is significantly different when combined with neural networks. Certain mitigations $-$ like delaying target network updates to control generalization and using estimates of expected TD errors in PER to avoid chasing stochasticity $-$ can avoid large spikes in error with PER and neural networks but generally do not outperform uniform replay. In control tasks, none of the prioritized variants consistently outperform uniform replay. We present new insight into the interaction between prioritization, bootstrapping, and neural networks and propose several improvements for PER in tabular settings and noisy domains.
Authors: Syrine Belakaria, Benjamin Letham, Janardhan Rao Doppa, Barbara Engelhardt, Stefano Ermon, Eytan Bakshy
Abstract: We consider the problem of active learning for global sensitivity analysis of expensive black-box functions. Our aim is to efficiently learn the importance of different input variables, e.g., in vehicle safety experimentation, we study the impact of the thickness of various components on safety objectives. Since function evaluations are expensive, we use active learning to prioritize experimental resources where they yield the most value. We propose novel active learning acquisition functions that directly target key quantities of derivative-based global sensitivity measures (DGSMs) under Gaussian process surrogate models. We showcase the first application of active learning directly to DGSMs, and develop tractable uncertainty reduction and information gain acquisition functions for these measures. Through comprehensive evaluation on synthetic and real-world problems, our study demonstrates how these active learning acquisition strategies substantially enhance the sample efficiency of DGSM estimation, particularly with limited evaluation budgets. Our work paves the way for more efficient and accurate sensitivity analysis in various scientific and engineering applications.
Authors: To Eun Kim, Alireza Salemi, Andrew Drozdov, Fernando Diaz, Hamed Zamani
Abstract: In the field of language modeling, models augmented with retrieval components have emerged as a promising solution to address several challenges faced in the natural language processing (NLP) field, including knowledge grounding, interpretability, and scalability. Despite the primary focus on NLP, we posit that the paradigm of retrieval-enhancement can be extended to a broader spectrum of machine learning (ML) such as computer vision, time series prediction, and computational biology. Therefore, this work introduces a formal framework of this paradigm, Retrieval-Enhanced Machine Learning (REML), by synthesizing the literature in various domains in ML with consistent notations which is missing from the current literature. Also, we found that while a number of studies employ retrieval components to augment their models, there is a lack of integration with foundational Information Retrieval (IR) research. We bridge this gap between the seminal IR research and contemporary REML studies by investigating each component that comprises the REML framework. Ultimately, the goal of this work is to equip researchers across various disciplines with a comprehensive, formally structured framework of retrieval-enhanced models, thereby fostering interdisciplinary future research.
Authors: Ye Lin Tun, Chu Myaet Thwal, Minh N. H. Nguyen, Choong Seon Hong
Abstract: Combining different data modalities enables deep neural networks to tackle complex tasks more effectively, making multimodal learning increasingly popular. To harness multimodal data closer to end users, it is essential to integrate multimodal learning with privacy-preserving approaches like federated learning (FL). However, compared to conventional unimodal learning, multimodal setting requires dedicated encoders for each modality, resulting in larger and more complex models. Training these models requires significant resources, presenting a substantial challenge for FL clients operating with limited computation and communication resources. To address these challenges, we introduce LW-FedMML, a layer-wise federated multimodal learning approach which decomposes the training process into multiple stages. Each stage focuses on training only a portion of the model, thereby significantly reducing the memory and computational requirements. Moreover, FL clients only need to exchange the trained model portion with the central server, lowering the resulting communication cost. We conduct extensive experiments across various FL and multimodal learning settings to validate the effectiveness of our proposed method. The results demonstrate that LW-FedMML can compete with conventional end-to-end federated multimodal learning (FedMML) while significantly reducing the resource burden on FL clients. Specifically, LW-FedMML reduces memory usage by up to $2.7\times$, computational operations (FLOPs) by $2.4\times$, and total communication cost by $2.3\times$. We also explore a progressive training approach called Prog-FedMML. While it offers lesser resource efficiency than LW-FedMML, Prog-FedMML has the potential to surpass the performance of end-to-end FedMML, making it a viable option for scenarios with fewer resource constraints.
Authors: Inkee Jung, Siu-Cheong Lau
Abstract: We present a local-to-global and measure-theoretical approach to understanding datasets. The core idea is to formulate a logifold structure and to interpret network models with restricted domains as local charts of datasets. In particular, this provides a mathematical foundation for ensemble machine learning. Our experiments demonstrate that logifolds can be implemented to identify fuzzy domains and improve accuracy compared to taking average of model outputs. Additionally, we provide a theoretical example of a logifold, highlighting the importance of restricting to domains of classifiers in an ensemble.
Authors: Serina Chang, Zhiyin Lin, Benjamin Yan, Swapnil Bembde, Qi Xiu, Chi Heem Wong, Yu Qin, Frank Kloster, Alex Luo, Raj Palleti, Jure Leskovec
Abstract: The global economy relies on the flow of goods over supply chain networks, with nodes as firms and edges as transactions between firms. While we may observe these external transactions, they are governed by unseen production functions, which determine how firms internally transform the input products they receive into output products that they sell. In this setting, it can be extremely valuable to infer these production functions, to improve supply chain visibility and to forecast future transactions more accurately. However, existing graph neural networks (GNNs) cannot capture these hidden relationships between nodes' inputs and outputs. Here, we introduce a new class of models for this setting by combining temporal GNNs with a novel inventory module, which learns production functions via attention weights and a special loss function. We evaluate our models extensively on real supply chains data and data generated from our new open-source simulator, SupplySim. Our models successfully infer production functions, outperforming the strongest baseline by 6%-50% (across datasets), and forecast future transactions, outperforming the strongest baseline by 11%-62%
Authors: Bo Wang, Wenbo Zhang, Yunpeng LI
Abstract: The elaborate pavement performance prediction is an important premise of implementing preventive maintenance. Our survey reveals that in practice, the pavement performance is usually measured at segment-level, where an unique performance value is obtained for all lanes within one segment of 1km length. It still lacks more elaborate performance analysis at lane-level due to costly data collection and difficulty in prediction modeling. Therefore, this study developed a multi-task deep learning approach to predict the lane-level pavement performance with a large amount of historical segment-level performance measurement data. The unified prediction framework can effectively address inherent correlation and differences across lanes. In specific, the prediction framework firstly employed an Long Short-Term Memory (LSTM) layer to capture the segment-level pavement deterioration pattern. Then multiple task-specific LSTM layers were designed based on number of lanes to capture lane-level differences in pavement performance. Finally, we concatenated multiple task-specific LSTM outputs with auxiliary features for further training and obtained the lane-level predictions after fully connected layer. The aforementioned prediction framework was validated with a real case in China. It revealed a better model performance regardless of one-way 2-lane, 3-lane, and 4-lane scenarios, all lower than 10% in terms of mean absolute percentage error. The proposed prediction framework also outperforms other ensemble learning and shallow machine learning methods in almost every lane.
Authors: Avisek Naug, Antonio Guillen, Ricardo Luna, Vineet Gundecha, Desik Rengarajan, Sahand Ghorbanpour, Sajad Mousavi, Ashwin Ramesh Babu, Dejan Markovikj, Lekhapriya D Kashyap, Soumyendu Sarkar
Abstract: Machine learning has driven an exponential increase in computational demand, leading to massive data centers that consume significant amounts of energy and contribute to climate change. This makes sustainable data center control a priority. In this paper, we introduce SustainDC, a set of Python environments for benchmarking multi-agent reinforcement learning (MARL) algorithms for data centers (DC). SustainDC supports custom DC configurations and tasks such as workload scheduling, cooling optimization, and auxiliary battery management, with multiple agents managing these operations while accounting for the effects of each other. We evaluate various MARL algorithms on SustainDC, showing their performance across diverse DC designs, locations, weather conditions, grid carbon intensity, and workload requirements. Our results highlight significant opportunities for improvement of data center operations using MARL algorithms. Given the increasing use of DC due to AI, SustainDC provides a crucial platform for the development and benchmarking of advanced algorithms essential for achieving sustainable computing and addressing other heterogeneous real-world challenges.
Authors: Venkatesh Balavadhani Parthasarathy, Ahtsham Zafar, Aafaq Khan, Arsalan Shahid
Abstract: This report examines the fine-tuning of Large Language Models (LLMs), integrating theoretical insights with practical applications. It outlines the historical evolution of LLMs from traditional Natural Language Processing (NLP) models to their pivotal role in AI. A comparison of fine-tuning methodologies, including supervised, unsupervised, and instruction-based approaches, highlights their applicability to different tasks. The report introduces a structured seven-stage pipeline for fine-tuning LLMs, spanning data preparation, model initialization, hyperparameter tuning, and model deployment. Emphasis is placed on managing imbalanced datasets and optimization techniques. Parameter-efficient methods like Low-Rank Adaptation (LoRA) and Half Fine-Tuning are explored for balancing computational efficiency with performance. Advanced techniques such as memory fine-tuning, Mixture of Experts (MoE), and Mixture of Agents (MoA) are discussed for leveraging specialized networks and multi-agent collaboration. The report also examines novel approaches like Proximal Policy Optimization (PPO) and Direct Preference Optimization (DPO), which align LLMs with human preferences, alongside pruning and routing optimizations to improve efficiency. Further sections cover validation frameworks, post-deployment monitoring, and inference optimization, with attention to deploying LLMs on distributed and cloud-based platforms. Emerging areas such as multimodal LLMs, fine-tuning for audio and speech, and challenges related to scalability, privacy, and accountability are also addressed. This report offers actionable insights for researchers and practitioners navigating LLM fine-tuning in an evolving landscape.
Authors: Yuxia Wu, Shujie Li, Yuan Fang, Chuan Shi
Abstract: Large language models (LLMs) have presented significant opportunities to enhance various machine learning applications, including graph neural networks (GNNs). By leveraging the vast open-world knowledge within LLMs, we can more effectively interpret and utilize textual data to better characterize heterophilic graphs, where neighboring nodes often have different labels. However, existing approaches for heterophilic graphs overlook the rich textual data associated with nodes, which could unlock deeper insights into their heterophilic contexts. In this work, we explore the potential of LLMs for modeling heterophilic graphs and propose a novel two-stage framework: LLM-enhanced edge discriminator and LLM-guided edge reweighting. In the first stage, we fine-tune the LLM to better identify homophilic and heterophilic edges based on the textual content of their nodes. In the second stage, we adaptively manage message propagation in GNNs for different edge types based on node features, structures, and heterophilic or homophilic characteristics. To cope with the computational demands when deploying LLMs in practical scenarios, we further explore model distillation techniques to fine-tune smaller, more efficient models that maintain competitive performance. Extensive experiments validate the effectiveness of our framework, demonstrating the feasibility of using LLMs to enhance node classification on heterophilic graphs.
Authors: Dorian Gailhard, Enzo Tartaglione, Lirida Naviner, Jhony H. Giraldo
Abstract: Hypergraphs are powerful mathematical structures that can model complex, high-order relationships in various domains, including social networks, bioinformatics, and recommender systems. However, generating realistic and diverse hypergraphs remains challenging due to their inherent complexity and lack of effective generative models. In this paper, we introduce a diffusion-based Hypergraph Generation (HYGENE) method that addresses these challenges through a progressive local expansion approach. HYGENE works on the bipartite representation of hypergraphs, starting with a single pair of connected nodes and iteratively expanding it to form the target hypergraph. At each step, nodes and hyperedges are added in a localized manner using a denoising diffusion process, which allows for the construction of the global structure before refining local details. Our experiments demonstrated the effectiveness of HYGENE, proving its ability to closely mimic a variety of properties in hypergraphs. To the best of our knowledge, this is the first attempt to employ deep learning models for hypergraph generation, and our work aims to lay the groundwork for future research in this area.
Authors: Emadeldeen Hamdan, Hongyi Pan, Ahmet Enis Cetin
Abstract: In this work, we introduce the concept of controllability and observability to the Mamba SSM's architecture in our Sparse-Mamba (S-Mamba) for natural language processing (NLP) applications. The structured state space model (SSM) development in recent studies, such as Mamba and Mamba2, outperformed and solved the computational inefficiency of transformers and large language models at small to medium scale. The Mamba SSMs architecture drops the need for attention layers or multilayer perception blocks in transformers. However, current Mamba models lack reinforcement of controllability in state-space equations for computing the $A$, $B$, $C$, and $D$ matrices at each time step, leading to increased complexity and computational costs. In this paper, we demonstrate a reduction of parameters in comparison to the first published Mamba and Mamba2. We showcase an improvement in perplexity by 5\% and a decrease in training time by 3\% after reinforcing controllability and observability on the original Mamba architecture in our proposed S-Mamba. The controllable $n \times n$ state matrix $A$ is sparse and it has only $n$ free parameters. Our novel approach will ensure a controllable system which will be the gate key for Mamba3.
Authors: Ramon Tavares, Ricardo Olinda
Abstract: This study presents a comprehensive methodology for modeling and forecasting the historical time series of active fire spots detected by the AQUA\_M-T satellite in the Amazon, Brazil. The approach employs a mixed Recurrent Neural Network (RNN) model, combining Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) architectures to predict the monthly accumulations of daily detected active fire spots. Data analysis revealed a consistent seasonality over time, with annual maximum and minimum values tending to repeat at the same periods each year. The primary objective is to verify whether the forecasts capture this inherent seasonality through machine learning techniques. The methodology involved careful data preparation, model configuration, and training using cross-validation with two seeds, ensuring that the data generalizes well to both the test and validation sets for both seeds. The results indicate that the combined LSTM and GRU model delivers excellent forecasting performance, demonstrating its effectiveness in capturing complex temporal patterns and modeling the observed time series. This research significantly contributes to the application of deep learning techniques in environmental monitoring, specifically in forecasting active fire spots. The proposed approach highlights the potential for adaptation to other time series forecasting challenges, opening new opportunities for research and development in machine learning and prediction of natural phenomena. Keywords: Time Series Forecasting; Recurrent Neural Networks; Deep Learning.
Authors: Evan Wang, Federico Cassano, Catherine Wu, Yunfeng Bai, Will Song, Vaskar Nath, Ziwen Han, Sean Hendryx, Summer Yue, Hugh Zhang
Abstract: While scaling training compute has led to remarkable improvements in large language models (LLMs), scaling inference compute has not yet yielded analogous gains. We hypothesize that a core missing component is a lack of diverse LLM outputs, leading to inefficient search due to models repeatedly sampling highly similar, yet incorrect generations. We empirically demonstrate that this lack of diversity can be mitigated by searching over candidate plans for solving a problem in natural language. Based on this insight, we propose PlanSearch, a novel search algorithm which shows strong results across HumanEval+, MBPP+, and LiveCodeBench (a contamination-free benchmark for competitive coding). PlanSearch generates a diverse set of observations about the problem and then uses these observations to construct plans for solving the problem. By searching over plans in natural language rather than directly over code solutions, PlanSearch explores a significantly more diverse range of potential solutions compared to baseline search methods. Using PlanSearch on top of Claude 3.5 Sonnet achieves a state-of-the-art pass@200 of 77.0% on LiveCodeBench, outperforming both the best score achieved without search (pass@1 = 41.4%) and using standard repeated sampling (pass@200 = 60.6%). Finally, we show that, across all models, search algorithms, and benchmarks analyzed, we can accurately predict performance gains due to search as a direct function of the diversity over generated ideas. Code can be found at https://github.com/scaleapi/plansearch.
Authors: Gangda Deng, Hongkuan Zhou, Rajgopal Kannan, Viktor Prasanna
Abstract: Heterophilous graphs, where dissimilar nodes tend to connect, pose a challenge for graph neural networks (GNNs) as their superior performance typically comes from aggregating homophilous information. Increasing the GNN depth can expand the scope (i.e., receptive field), potentially finding homophily from the higher-order neighborhoods. However, uniformly expanding the scope results in subpar performance since real-world web graphs often exhibit homophily disparity between nodes. An ideal way is personalized scopes, allowing nodes to have varying scope sizes. Existing methods typically add node-adaptive weights for each hop. Although expressive, they inevitably suffer from severe overfitting. To address this issue, we formalize personalized scoping as a separate scope classification problem that overcomes GNN overfitting in node classification. Specifically, we predict the optimal GNN depth for each node. Our theoretical and empirical analysis suggests that accurately predicting the depth can significantly enhance generalization. We further propose Adaptive Scope (AS), a lightweight approach that only participates in GNN inference. AS encodes structural patterns and predicts the depth to select the best model for each node's prediction. Experimental results show that AS is highly flexible with various GNN architectures across a wide range of datasets while significantly improving accuracy.
Authors: Luning Wang, Shiyao Li, Xuefei Ning, Zhihang Yuan, Shengen Yan, Guohao Dai, Yu Wang
Abstract: Large Language Models (LLMs) have been widely adopted to process long-context tasks. However, the large memory overhead of the key-value (KV) cache poses significant challenges in long-context scenarios. Existing training-free KV cache compression methods typically focus on quantization and token pruning, which have compression limits, and excessive sparsity can lead to severe performance degradation. Other methods design new architectures with less KV overhead but require significant training overhead. To address the above two drawbacks, we further explore the redundancy in the channel dimension and apply an architecture-level design with minor training costs. Therefore, we introduce CSKV, a training-efficient Channel Shrinking technique for KV cache compression: (1) We first analyze the singular value distribution of the KV cache, revealing significant redundancy and compression potential along the channel dimension. Based on this observation, we propose using low-rank decomposition for key and value layers and storing the low-dimension features. (2) To preserve model performance, we introduce a bi-branch KV cache, including a window-based full-precision KV cache and a low-precision compressed KV cache. (3) To reduce the training costs, we minimize the layer-wise reconstruction loss for the compressed KV cache instead of retraining the entire LLMs. Extensive experiments show that CSKV can reduce the memory overhead of the KV cache by 80% while maintaining the model's long-context capability. Moreover, we show that our method can be seamlessly combined with quantization to further reduce the memory overhead, achieving a compression ratio of up to 95%. Code is available at https://github.com/wln20/CSKV.
Authors: Yifan Tan, Haoze Wang, Chao Yan, Yangdong Deng
Abstract: Model quantization has become a crucial technique to address the issues of large memory consumption and long inference times associated with LLMs. Mixed-precision quantization, which distinguishes between important and unimportant parameters, stands out among numerous quantization schemes as it achieves a balance between precision and compression rate. However, existing approaches can only identify important parameters through qualitative analysis and manual experiments without quantitatively analyzing how their importance is determined. We propose a new criterion, so-called 'precision alignment', to build a quantitative framework to holistically evaluate the importance of parameters in mixed-precision quantization. Our observations on floating point addition under various real-world scenarios suggest that two addends should have identical precision, otherwise the information in the higher-precision number will be wasted. Such an observation offers an essential principle to determine the precision of each parameter in matrix multiplication operation. As the first step towards applying the above discovery to large model inference, we develop a dynamic KV-Cache quantization technique to effectively reduce memory access latency. Different from existing quantization approaches that focus on memory saving, this work directly aims to accelerate LLM inference through quantifying floating numbers. The proposed technique attains a 25% saving of memory access and delivers up to 1.3x speedup in the computation of attention in the decoding phase of LLM, with almost no loss of precision.
Authors: Michael S. Albergo, Eric Vanden-Eijnden
Abstract: We propose an algorithm, termed the Non-Equilibrium Transport Sampler (NETS), to sample from unnormalized probability distributions. NETS can be viewed as a variant of annealed importance sampling (AIS) based on Jarzynski's equality, in which the stochastic differential equation used to perform the non-equilibrium sampling is augmented with an additional learned drift term that lowers the impact of the unbiasing weights used in AIS. We show that this drift is the minimizer of a variety of objective functions, which can all be estimated in an unbiased fashion without backpropagating through solutions of the stochastic differential equations governing the sampling. We also prove that some these objectives control the Kullback-Leibler divergence of the estimated distribution from its target. NETS is shown to be unbiased and, in addition, has a tunable diffusion coefficient which can be adjusted post-training to maximize the effective sample size. We demonstrate the efficacy of the method on standard benchmarks, high-dimensional Gaussian mixture distributions, and a model from statistical lattice field theory, for which it surpasses the performances of related work and existing baselines.
Authors: Jun Li, Aaron Aguirre, Junior Moura, Che Liu, Lanhai Zhong, Chenxi Sun, Gari Clifford, Brandon Westover, Shenda Hong
Abstract: Artificial intelligence (AI) has demonstrated significant potential in ECG analysis and cardiovascular disease assessment. Recently, foundation models have played a remarkable role in advancing medical AI. The development of an ECG foundation model holds the promise of elevating AI-ECG research to new heights. However, building such a model faces several challenges, including insufficient database sample sizes and inadequate generalization across multiple domains. Additionally, there is a notable performance gap between single-lead and multi-lead ECG analyses. We introduced an ECG Foundation Model (ECGFounder), a general-purpose model that leverages real-world ECG annotations from cardiology experts to broaden the diagnostic capabilities of ECG analysis. ECGFounder was trained on over 10 million ECGs with 150 label categories from the Harvard-Emory ECG Database, enabling comprehensive cardiovascular disease diagnosis through ECG analysis. The model is designed to be both an effective out-of-the-box solution, and a to be fine-tunable for downstream tasks, maximizing usability. Importantly, we extended its application to lower rank ECGs, and arbitrary single-lead ECGs in particular. ECGFounder is applicable to supporting various downstream tasks in mobile monitoring scenarios. Experimental results demonstrate that ECGFounder achieves expert-level performance on internal validation sets, with AUROC exceeding 0.95 for eighty diagnoses. It also shows strong classification performance and generalization across various diagnoses on external validation sets. When fine-tuned, ECGFounder outperforms baseline models in demographic analysis, clinical event detection, and cross-modality cardiac rhythm diagnosis. The trained model and data will be publicly released upon publication through the bdsp.io. Our code is available at https://github.com/bdsp-core/ECGFounder
Authors: Ruinan Jin, Xiao Li, Yaoliang Yu, Baoxiang Wang
Abstract: Adaptive Moment Estimation (Adam) is a cornerstone optimization algorithm in deep learning, widely recognized for its flexibility with adaptive learning rates and efficiency in handling large-scale data. However, despite its practical success, the theoretical understanding of Adam's convergence has been constrained by stringent assumptions, such as almost surely bounded stochastic gradients or uniformly bounded gradients, which are more restrictive than those typically required for analyzing stochastic gradient descent (SGD). In this paper, we introduce a novel and comprehensive framework for analyzing the convergence properties of Adam. This framework offers a versatile approach to establishing Adam's convergence. Specifically, we prove that Adam achieves asymptotic (last iterate sense) convergence in both the almost sure sense and the \(L_1\) sense under the relaxed assumptions typically used for SGD, namely \(L\)-smoothness and the ABC inequality. Meanwhile, under the same assumptions, we show that Adam attains non-asymptotic sample complexity bounds similar to those of SGD.
Authors: Alex Iacob, Lorenzo Sani, Meghdad Kurmanji, William F. Shen, Xinchi Qiu, Dongqi Cai, Yan Gao, Nicholas D. Lane
Abstract: Language model pre-training benefits from diverse data to enhance performance across domains and languages. However, training on such heterogeneous corpora requires extensive and costly efforts. Since these data sources vary lexically, syntactically, and semantically, they cause negative interference or the ``curse of multilinguality''. We propose a novel pre-training framework to alleviate this curse. Our method, DEPT, decouples embeddings from the transformer body while simultaneously training the latter in multiple contexts. DEPT enables training without a shared global vocabulary and: (1) can train robustly and effectively under significant data heterogeneity, (2) reduces token embedding parameters by up to 80% and the communication costs by 675x for billion-scale models, (3) enhances model generalization and plasticity in adapting to new languages and domains, and (4) permits training with custom optimized vocabularies per data source. We demonstrate DEPT's potential via the first vocabulary-agnostic federated multilingual pre-training of a 1.3 billion-parameter model, limiting its embedding size to 102.4 million instead of 512 million.
Authors: Chuansen Peng, Hanning Tang, Zhiguo Wang, Xiaojing Shen
Abstract: Inferring network topology from smooth signals is a significant problem in data science and engineering. A common challenge in real-world scenarios is the availability of only partially observed nodes. While some studies have considered hidden nodes and proposed various optimization frameworks, existing methods often lack the practical efficiency needed for large-scale networks or fail to provide theoretical convergence guarantees. In this paper, we address the problem of inferring network topologies from smooth signals with partially observed nodes. We propose a first-order algorithmic framework that includes two variants: one based on column sparsity regularization and the other on a low-rank constraint. We establish theoretical convergence guarantees and demonstrate the linear convergence rate of our algorithms. Extensive experiments on both synthetic and real-world data show that our results align with theoretical predictions, exhibiting not only linear convergence but also superior speed compared to existing methods. To the best of our knowledge, this is the first work to propose a first-order algorithmic framework for inferring network structures from smooth signals under partial observability, offering both guaranteed linear convergence and practical effectiveness for large-scale networks.
Authors: Wei Liu, Zhiying Deng, Zhongyu Niu, Jun Wang, Haozhao Wang, YuanKai Zhang, Ruixuan Li
Abstract: An important line of research in the field of explainability is to extract a small subset of crucial rationales from the full input. The most widely used criterion for rationale extraction is the maximum mutual information (MMI) criterion. However, in certain datasets, there are spurious features non-causally correlated with the label and also get high mutual information, complicating the loss landscape of MMI. Although some penalty-based methods have been developed to penalize the spurious features (e.g., invariance penalty, intervention penalty, etc) to help MMI work better, these are merely remedial measures. In the optimization objectives of these methods, spurious features are still distinguished from plain noise, which hinders the discovery of causal rationales. This paper aims to develop a new criterion that treats spurious features as plain noise, allowing the model to work on datasets rich in spurious features as if it were working on clean datasets, thereby making rationale extraction easier. We theoretically observe that removing either plain noise or spurious features from the input does not alter the conditional distribution of the remaining components relative to the task label. However, significant changes in the conditional distribution occur only when causal features are eliminated. Based on this discovery, the paper proposes a criterion for \textbf{M}aximizing the \textbf{R}emaining \textbf{D}iscrepancy (MRD). Experiments on six widely used datasets show that our MRD criterion improves rationale quality (measured by the overlap with human-annotated rationales) by up to $10.4\%$ as compared to several recent competitive MMI variants. Code: \url{https://github.com/jugechengzi/Rationalization-MRD}.
Authors: Junjie Chen, Qian Chen, Jian Lou, Xiaoyu Zhang, Kai Wu, Zilong Wang
Abstract: Machine unlearning (MU) is becoming a promising paradigm to achieve the "right to be forgotten", where the training trace of any chosen data points could be eliminated, while maintaining the model utility on general testing samples after unlearning. With the advancement of forgetting research, many fundamental open questions remain unanswered: do different samples exhibit varying levels of difficulty in being forgotten? Further, does the sequence in which samples are forgotten, determined by their respective difficulty levels, influence the performance of forgetting algorithms? In this paper, we identify key factor affecting unlearning difficulty and the performance of unlearning algorithms. We find that samples with higher privacy risks are more likely to be unlearning, indicating that the unlearning difficulty varies among different samples which motives a more precise unlearning mode. Built upon this insight, we propose a general unlearning framework, dubbed RSU, which consists of Ranking module and SeqUnlearn module.
Authors: Sagi Shaier, Francisco Pereira, Katharina von der Wense, Lawrence E Hunter, Matt Jones
Abstract: The evolution of biological neural systems has led to both modularity and sparse coding, which enables efficiency in energy usage, and robustness across the diversity of tasks in the lifespan. In contrast, standard neural networks rely on dense, non-specialized architectures, where all model parameters are simultaneously updated to learn multiple tasks, leading to representation interference. Current sparse neural network approaches aim to alleviate this issue, but are often hindered by limitations such as 1) trainable gating functions that cause representation collapse; 2) non-overlapping experts that result in redundant computation and slow learning; and 3) reliance on explicit input or task IDs that impose significant constraints on flexibility and scalability. In this paper we propose Conditionally Overlapping Mixture of ExperTs (COMET), a general deep learning method that addresses these challenges by inducing a modular, sparse architecture with an exponential number of overlapping experts. COMET replaces the trainable gating function used in Sparse Mixture of Experts with a fixed, biologically inspired random projection applied to individual input representations. This design causes the degree of expert overlap to depend on input similarity, so that similar inputs tend to share more parameters. This facilitates positive knowledge transfer, resulting in faster learning and improved generalization. We demonstrate the effectiveness of COMET on a range of tasks, including image classification, language modeling, and regression, using several popular deep learning architectures.
Authors: Raphael C. Kim, Falco J. Bargagli-Stoffi, Kevin L. Chen, Rachel C. Nethery
Abstract: The substantial effect of air pollution on cardiovascular disease and mortality burdens is well-established. Emissions-reducing interventions on coal-fired power plants -- a major source of hazardous air pollution -- have proven to be an effective, but costly, strategy for reducing pollution-related health burdens. Targeting the power plants that achieve maximum health benefits while satisfying realistic cost constraints is challenging. The primary difficulty lies in quantifying the health benefits of intervening at particular plants. This is further complicated because interventions are applied on power plants, while health impacts occur in potentially distant communities, a setting known as bipartite network interference (BNI). In this paper, we introduce novel policy learning methods based on Q- and A-Learning to determine the optimal policy under arbitrary BNI. We derive asymptotic properties and demonstrate finite sample efficacy in simulations. We apply our novel methods to a comprehensive dataset of Medicare claims, power plant data, and pollution transport networks. Our goal is to determine the optimal strategy for installing power plant scrubbers to minimize ischemic heart disease (IHD) hospitalizations under various cost constraints. We find that annual IHD hospitalization rates could be reduced in a range from 20.66-44.51 per 10,000 person-years through optimal policies under different cost constraints.
Authors: Zhenghao Xu, Yuqing Wang, Tuo Zhao, Rachel Ward, Molei Tao
Abstract: We study the convergence rate of first-order methods for rectangular matrix factorization, which is a canonical nonconvex optimization problem. Specifically, given a rank-$r$ matrix $\mathbf{A}\in\mathbb{R}^{m\times n}$, we prove that gradient descent (GD) can find a pair of $\epsilon$-optimal solutions $\mathbf{X}_T\in\mathbb{R}^{m\times d}$ and $\mathbf{Y}_T\in\mathbb{R}^{n\times d}$, where $d\geq r$, satisfying $\lVert\mathbf{X}_T\mathbf{Y}_T^\top-\mathbf{A}\rVert_\mathrm{F}\leq\epsilon\lVert\mathbf{A}\rVert_\mathrm{F}$ in $T=O(\kappa^2\log\frac{1}{\epsilon})$ iterations with high probability, where $\kappa$ denotes the condition number of $\mathbf{A}$. Furthermore, we prove that Nesterov's accelerated gradient (NAG) attains an iteration complexity of $O(\kappa\log\frac{1}{\epsilon})$, which is the best-known bound of first-order methods for rectangular matrix factorization. Different from small balanced random initialization in the existing literature, we adopt an unbalanced initialization, where $\mathbf{X}_0$ is large and $\mathbf{Y}_0$ is $0$. Moreover, our initialization and analysis can be further extended to linear neural networks, where we prove that NAG can also attain an accelerated linear convergence rate. In particular, we only require the width of the network to be greater than or equal to the rank of the output label matrix. In contrast, previous results achieving the same rate require excessive widths that additionally depend on the condition number and the rank of the input data matrix.
Authors: Dan Ley, Suraj Srinivas, Shichang Zhang, Gili Rusak, Himabindu Lakkaraju
Abstract: Data Attribution (DA) methods quantify the influence of individual training data points on model outputs and have broad applications such as explainability, data selection, and noisy label identification. However, existing DA methods are often computationally intensive, limiting their applicability to large-scale machine learning models. To address this challenge, we introduce the Generalized Group Data Attribution (GGDA) framework, which computationally simplifies DA by attributing to groups of training points instead of individual ones. GGDA is a general framework that subsumes existing attribution methods and can be applied to new DA techniques as they emerge. It allows users to optimize the trade-off between efficiency and fidelity based on their needs. Our empirical results demonstrate that GGDA applied to popular DA methods such as Influence Functions, TracIn, and TRAK results in upto 10x-50x speedups over standard DA methods while gracefully trading off attribution fidelity. For downstream applications such as dataset pruning and noisy label identification, we demonstrate that GGDA significantly improves computational efficiency and maintains effectiveness, enabling practical applications in large-scale machine learning scenarios that were previously infeasible.
Authors: Maham Ilyas, Abida Yasmeen, Yaser Daanial Khan, Arif Mahmood
Abstract: Acetylation of lysine residues (K-Ace) is a post-translation modification occurring in both prokaryotes and eukaryotes. It plays a crucial role in disease pathology and cell biology hence it is important to identify these K-Ace sites. In the past, many machine learning-based models using hand-crafted features and encodings have been used to find and analyze the characteristics of K-Ace sites however these methods ignore long term relationships within sequences and therefore observe performance degradation. In the current work we propose Deep-Ace, a deep learning-based framework using Long-Short-Term-Memory (LSTM) network which has the ability to understand and encode long-term relationships within a sequence. Such relations are vital for learning discriminative and effective sequence representations. In the work reported here, the use of LSTM to extract deep features as well as for prediction of K-Ace sites using fully connected layers for eight different species of prokaryotic models (including B. subtilis, C. glutamicum, E. coli, G. kaustophilus, S. eriocheiris, B. velezensis, S. typhimurium, and M. tuberculosis) has been explored. Our proposed method has outperformed existing state of the art models achieving accuracy as 0.80, 0.79, 0.71, 0.75, 0.80, 0.83, 0.756, and 0.82 respectively for eight bacterial species mentioned above. The method with minor modifications can be used for eukaryotic systems and can serve as a tool for the prognosis and diagnosis of various diseases in humans.
Authors: Andrew Gracyk
Abstract: We develop Riemannian approaches to variational autoencoders (VAEs) for PDE-type ambient data with regularizing geometric latent dynamics, which we refer to as VAE-DLM, or VAEs with dynamical latent manifolds. We redevelop the VAE framework such that manifold geometries, subject to our geometric flow, embedded in Euclidean space are learned in the intermediary latent space developed by encoders and decoders. By tailoring the geometric flow in which the latent space evolves, we induce latent geometric properties of our choosing, which are reflected in empirical performance. We reformulate the traditional evidence lower bound (ELBO) loss with a considerate choice of prior. We develop a linear geometric flow with a steady-state regularizing term. This flow requires only automatic differentiation of one time derivative, and can be solved in moderately high dimensions in a physics-informed approach, allowing more expressive latent representations. We discuss how this flow can be formulated as a gradient flow, and maintains entropy away from metric singularity. This, along with an eigenvalue penalization condition, helps ensure the manifold is sufficiently large in measure, nondegenerate, and a canonical geometry, which contribute to a robust representation. Our methods focus on the modified multi-layer perceptron architecture with tanh activations for the manifold encoder-decoder. We demonstrate, on our datasets of interest, our methods perform at least as well as the traditional VAE, and oftentimes better. Our methods can outperform this and a VAE endowed with our proposed architecture by up to 25% reduction in out-of-distribution (OOD) error and potentially greater. We highlight our method on ambient PDEs whose solutions maintain minimal variation in late times. We provide empirical justification towards how we can improve robust learning for external dynamics with VAEs.
Authors: Junkang Wu, Xue Wang, Zhengyi Yang, Jiancan Wu, Jinyang Gao, Bolin Ding, Xiang Wang, Xiangnan He
Abstract: Aligning large language models (LLMs) with human values and intentions is crucial for their utility, honesty, and safety. Reinforcement learning from human feedback (RLHF) is a popular approach to achieve this alignment, but it faces challenges in computational efficiency and training stability. Recent methods like Direct Preference Optimization (DPO) and Simple Preference Optimization (SimPO) have proposed offline alternatives to RLHF, simplifying the process by reparameterizing the reward function. However, DPO depends on a potentially suboptimal reference model, and SimPO's assumption of a fixed target reward margin may lead to suboptimal decisions in diverse data settings. In this work, we propose $\alpha$-DPO, an adaptive preference optimization algorithm designed to address these limitations by introducing a dynamic reward margin. Specifically, $\alpha$-DPO employs an adaptive preference distribution, balancing the policy model and the reference model to achieve personalized reward margins. We provide theoretical guarantees for $\alpha$-DPO, demonstrating its effectiveness as a surrogate optimization objective and its ability to balance alignment and diversity through KL divergence control. Empirical evaluations on AlpacaEval 2 and Arena-Hard show that $\alpha$-DPO consistently outperforms DPO and SimPO across various model settings, establishing it as a robust approach for fine-tuning LLMs. Our method achieves significant improvements in win rates, highlighting its potential as a powerful tool for LLM alignment. The code is available at https://github.com/junkangwu/alpha-DPO
Authors: Frederiek Wesel, Kim Batselier
Abstract: The ability to express a learning task in terms of a primal and a dual optimization problem lies at the core of a plethora of machine learning methods. For example, Support Vector Machine (SVM), Least-Squares Support Vector Machine (LS-SVM), Ridge Regression (RR), Lasso Regression (LR), Principal Component Analysis (PCA), and more recently Singular Value Decomposition (SVD) have all been defined either in terms of primal weights or in terms of dual Lagrange multipliers. The primal formulation is computationally advantageous in the case of large sample size while the dual is preferred for high-dimensional data. Crucially, said learning problems can be made nonlinear through the introduction of a feature map in the primal problem, which corresponds to applying the kernel trick in the dual. In this paper we derive a primal-dual formulation of the Multilinear Singular Value Decomposition (MLSVD), which recovers as special cases both PCA and SVD. Besides enabling computational gains through the derived primal formulation, we propose a nonlinear extension of the MLSVD using feature maps, which results in a dual problem where a kernel tensor arises. We discuss potential applications in the context of signal analysis and deep learning.
Authors: Juan Sebastian Rojas, Chi-Guhn Lee
Abstract: Average-reward Markov decision processes (MDPs) provide a foundational framework for sequential decision-making under uncertainty. However, average-reward MDPs have remained largely unexplored in reinforcement learning (RL) settings, with the majority of RL-based efforts having been allocated to episodic and discounted MDPs. In this work, we study a unique structural property of average-reward MDPs and utilize it to introduce Reward-Extended Differential (or RED) reinforcement learning: a novel RL framework that can be used to effectively and efficiently solve various subtasks simultaneously in the average-reward setting. We introduce a family of RED learning algorithms for prediction and control, including proven-convergent algorithms for the tabular case. We then showcase the power of these algorithms by demonstrating how they can be used to learn a policy that optimizes, for the first time, the well-known conditional value-at-risk (CVaR) risk measure in a fully-online manner, without the use of an explicit bi-level optimization scheme or an augmented state-space.
Authors: Xurui Li, Juanjuan Yao
Abstract: The advent of Large Language Models (LLMs) has ushered in a new era of artificial intelligence, with the potential to transform various sectors through automation and insightful analysis. The Mixture of Experts (MoE) architecture has been proposed as a solution to enhance model performance in complex tasks. Yet, existing MoE models struggle with task-specific learning and interpretability, especially in fields like medicine where precision is critical. This paper introduces the Adaptive Task-planing Mixture of Experts(AT-MoE), an innovative architecture designed to address these limitations. We first train task-specific experts via LoRA approach to enhance problem-solving capabilities and interpretability in specialized areas. Subsequently, we introduce a layer-wise adaptive grouped routing module that optimizes module fusion based on complex task instructions, ensuring optimal task resolution. The grouped routing module first perform overall weight allocation from the dimension of the expert group, and then conduct local weight normalization adjustments within the group. This design maintains multi-dimensional balance, controllability, and interpretability, while facilitating task-specific fusion in response to complex instructions.
Authors: Christofel Rio Goenawan
Abstract: In the modern world, the development of Artificial Intelligence (AI) has contributed to improvements in various areas, including automation, computer vision, fraud detection, and more. AI can be leveraged to enhance the efficiency of Autonomous Smart Traffic Management (ASTM) systems and reduce traffic congestion rates. This paper presents an Autonomous Smart Traffic Management (STM) system that uses AI to improve traffic flow rates. The system employs the YOLO V5 Convolutional Neural Network to detect vehicles in traffic management images. Additionally, it predicts the number of vehicles for the next 12 hours using a Recurrent Neural Network with Long Short-Term Memory (RNN-LSTM). The Smart Traffic Management Cycle Length Analysis manages the traffic cycle length based on these vehicle predictions, aided by AI. From the results of the RNN-LSTM model for predicting vehicle numbers over the next 12 hours, we observe that the model predicts traffic with a Mean Squared Error (MSE) of 4.521 vehicles and a Root Mean Squared Error (RMSE) of 2.232 vehicles. After simulating the STM system in the CARLA simulation environment, we found that the Traffic Management Congestion Flow Rate with ASTM (21 vehicles per minute) is 50\% higher than the rate without STM (around 15 vehicles per minute). Additionally, the Traffic Management Vehicle Pass Delay with STM (5 seconds per vehicle) is 70\% lower than without STM (around 12 seconds per vehicle). These results demonstrate that the STM system using AI can increase traffic flow by 50\% and reduce vehicle pass delays by 70\%.
Authors: Sujai Hiremath, Promit Ghosal, Kyra Gan
Abstract: We address the challenge of causal discovery in structural equation models with additive noise without imposing additional assumptions on the underlying data-generating process. We introduce local search in additive noise model (LoSAM), which generalizes an existing nonlinear method that leverages local causal substructures to the general additive noise setting, allowing for both linear and nonlinear causal mechanisms. We show that LoSAM achieves polynomial runtime, and improves runtime and efficiency by exploiting new substructures to minimize the conditioning set at each step. Further, we introduce a variant of LoSAM, LoSAM-UC, that is robust to unmeasured confounding among roots, a property that is often not satisfied by functional-causal-model-based methods. We numerically demonstrate the utility of LoSAM, showing that it outperforms existing benchmarks.
Authors: Zhe Li, Xiangfei Qiu, Peng Chen, Yihang Wang, Hanyin Cheng, Yang Shu, Jilin Hu, Chenjuan Guo, Aoying Zhou, Qingsong Wen, Christian S. Jensen, Bin Yang
Abstract: Time Series Forecasting (TSF) is key functionality in numerous fields, including in finance, weather services, and energy management. While TSF methods are emerging these days, many of them require domain-specific data collection and model training and struggle with poor generalization performance on new domains. Foundation models aim to overcome this limitation. Pre-trained on large-scale language or time series data, they exhibit promising inferencing capabilities in new or unseen data. This has spurred a surge in new TSF foundation models. We propose a new benchmark, FoundTS, to enable thorough and fair evaluation and comparison of such models. FoundTS covers a variety of TSF foundation models, including those based on large language models and those pretrained on time series. Next, FoundTS supports different forecasting strategies, including zero-shot, few-shot, and full-shot, thereby facilitating more thorough evaluations. Finally, FoundTS offers a pipeline that standardizes evaluation processes such as dataset splitting, loading, normalization, and few-shot sampling, thereby facilitating fair evaluations. Building on this, we report on an extensive evaluation of TSF foundation models on a broad range of datasets from diverse domains and with different statistical characteristics. Specifically, we identify pros and cons and inherent limitations of existing foundation models, and we identify directions for future model design. We make our code and datasets available at https://anonymous.4open.science/r/FoundTS-C2B0.
Authors: Zeev Yampolsky, Itzik Klein
Abstract: Gyroscopes are inertial sensors that measure the angular velocity of the platforms to which they are attached. To estimate the gyroscope deterministic error terms prior mission start, a calibration procedure is performed. When considering low-cost gyroscopes, the calibration requires a turntable as the gyros are incapable of sensing the Earth turn rate. In this paper, we propose a data-driven framework to estimate the scale factor and bias of a gyroscope. To train and validate our approach, a dataset of 56 minutes was recorded using a turntable. We demonstrated that our proposed approach outperforms the model-based approach, in terms of accuracy and convergence time. Specifically, we improved the scale factor and bias estimation by an average of 72% during six seconds of calibration time, demonstrating an average of 75% calibration time improvement. That is, instead of minutes, our approach requires only several seconds for the calibration.
Authors: Ching Chang, Chiao-Tung Chan, Wei-Yao Wang, Wen-Chih Peng, Tien-Fu Chen
Abstract: Multivariate time-series data in fields like healthcare and industry are informative but challenging due to high dimensionality and lack of labels. Recent self-supervised learning methods excel in learning rich representations without labels but struggle with disentangled embeddings and inductive bias issues like transformation-invariance. To address these challenges, we introduce TimeDRL, a framework for multivariate time-series representation learning with dual-level disentangled embeddings. TimeDRL features: (i) disentangled timestamp-level and instance-level embeddings using a [CLS] token strategy; (ii) timestamp-predictive and instance-contrastive tasks for representation learning; and (iii) avoidance of augmentation methods to eliminate inductive biases. Experiments on forecasting and classification datasets show TimeDRL outperforms existing methods, with further validation in semi-supervised settings with limited labeled data.
Authors: Xingzhi Sun, Danqi Liao, Kincaid MacDonald, Yanlei Zhang, Chen Liu, Guillaume Huguet, Guy Wolf, Ian Adelstein, Tim G. J. Rudner, Smita Krishnaswamy
Abstract: Rapid growth of high-dimensional datasets in fields such as single-cell RNA sequencing and spatial genomics has led to unprecedented opportunities for scientific discovery, but it also presents unique computational and statistical challenges. Traditional methods struggle with geometry-aware data generation, interpolation along meaningful trajectories, and transporting populations via feasible paths. To address these issues, we introduce Geometry-Aware Generative Autoencoder (GAGA), a novel framework that combines extensible manifold learning with generative modeling. GAGA constructs a neural network embedding space that respects the intrinsic geometries discovered by manifold learning and learns a novel warped Riemannian metric on the data space. This warped metric is derived from both the points on the data manifold and negative samples off the manifold, allowing it to characterize a meaningful geometry across the entire latent space. Using this metric, GAGA can uniformly sample points on the manifold, generate points along geodesics, and interpolate between populations across the learned manifold using geodesic-guided flows. GAGA shows competitive performance in simulated and real-world datasets, including a 30% improvement over the state-of-the-art methods in single-cell population-level trajectory inference.
Authors: Khiem Le, Nitesh V. Chawla
Abstract: Molecule optimization is a critical task in drug discovery to optimize desired properties of a given molecule through chemical modification. Despite Large Language Models (LLMs) holding the potential to efficiently simulate this task by using natural language to direct the optimization, straightforwardly utilizing shows limited performance. In this work, we facilitate utilizing LLMs in an iterative paradigm by proposing a simple yet highly effective domain feedback provider, namely $\text{Re}^2$DF. In detail, $\text{Re}^2$DF harnesses an external toolkit, RDKit, to handle the molecule hallucination, if the modified molecule is chemically invalid. Otherwise, its desired properties are computed and compared to the original one, establishing reliable domain feedback with correct direction and distance towards the objective, followed by a retrieved example, to explicitly guide the LLM to refine the modified molecule. We conduct experiments across both single- and multi-property objectives with 2 thresholds, where $\text{Re}^2$DF shows significant improvements. Particularly, for 20 single-property objectives, $\text{Re}^2$DF enhances Hit ratio by 16.95% and 20.76% under loose and strict thresholds, respectively. For 32 multi-property objectives, $\text{Re}^2$DF enhances Hit ratio by 6.04% and 5.25%.
Authors: Al Zadid Sultan Bin Habib, Kesheng Wang, Mary-Anne Hartley, Gianfranco Doretto, Donald A. Adjeroh
Abstract: Effective analysis of tabular data still poses a significant problem in deep learning, mainly because features in tabular datasets are often heterogeneous and have different levels of relevance. This work introduces TabSeq, a novel framework for the sequential ordering of features, addressing the vital necessity to optimize the learning process. Features are not always equally informative, and for certain deep learning models, their random arrangement can hinder the model's learning capacity. Finding the optimum sequence order for such features could improve the deep learning models' learning process. The novel feature ordering technique we provide in this work is based on clustering and incorporates both local ordering and global ordering. It is designed to be used with a multi-head attention mechanism in a denoising autoencoder network. Our framework uses clustering to align comparable features and improve data organization. Multi-head attention focuses on essential characteristics, whereas the denoising autoencoder highlights important aspects by rebuilding from distorted inputs. This method improves the capability to learn from tabular data while lowering redundancy. Our research, demonstrating improved performance through appropriate feature sequence rearrangement using raw antibody microarray and two other real-world biomedical datasets, validates the impact of feature ordering. These results demonstrate that feature ordering can be a viable approach to improved deep learning of tabular data.
Authors: Alex Mallen, Nora Belrose
Abstract: Scalable oversight studies methods of training and evaluating AI systems in domains where human judgment is unreliable or expensive, such as scientific research and software engineering in complex codebases. Most work in this area has focused on methods of improving the quality of labels. Recent work by Burns et al. (2023) considers the complementary problem of training models with low-quality labels, finding that large pretrained models often have an inductive bias towards producing correct answers. In practice, however, neither label quantity nor quality is fixed: practitioners face a quantity-quality tradeoff. In this paper, we explore the microeconomics of the quantity-quality tradeoff on binary NLP classification tasks used in Burns et al. (2023). While sample-efficient learning has been studied extensively, little public research has focused on scalable elicitation: eliciting capabilities from pretrained models subject to labeling cost constraints. We find that this setting has novel dynamics caused by the tradeoff between label quantity and quality, as well as the model's existing latent capabilities. We observe three regimes of eliciting classification knowledge from pretrained models using supervised finetuning: quantity-dominant, quality-dominant, and a mixed regime involving the use of low- and high-quality data together to attain higher accuracy at a lower cost than using either alone. We explore sample-efficient elicitation methods that make use of two datasets of differing qualities, and establish a Pareto frontier of scalable elicitation methods that optimally trade off labeling cost and classifier performance. We find that the accuracy of supervised fine-tuning can be improved by up to 5 percentage points at a fixed labeling budget by adding a few-shot prompt to make use of the model's existing knowledge of the task.
Authors: Jake Robertson, Thorsten Schmidt, Frank Hutter, Noor Awad
Abstract: Fairness-aware Machine Learning (FairML) applications are often characterized by complex social objectives and legal requirements, frequently involving multiple, potentially conflicting notions of fairness. Despite the well-known Impossibility Theorem of Fairness and extensive theoretical research on the statistical and socio-technical trade-offs between fairness metrics, many FairML tools still optimize or constrain for a single fairness objective. However, this one-sided optimization can inadvertently lead to violations of other relevant notions of fairness. In this socio-technical and empirical study, we frame fairness as a many-objective (MaO) problem by treating fairness metrics as conflicting objectives. We introduce ManyFairHPO, a human-in-the-loop, fairness-aware model selection framework that enables practitioners to effectively navigate complex and nuanced fairness objective landscapes. ManyFairHPO aids in the identification, evaluation, and balancing of fairness metric conflicts and their related social consequences, leading to more informed and socially responsible model-selection decisions. Through a comprehensive empirical evaluation and a case study on the Law School Admissions problem, we demonstrate the effectiveness of ManyFairHPO in balancing multiple fairness objectives, mitigating risks such as self-fulfilling prophecies, and providing interpretable insights to guide stakeholders in making fairness-aware modeling decisions.
Authors: Andreas Opedal, Haruki Shirakami, Bernhard Sch\"olkopf, Abulhair Saparov, Mrinmaya Sachan
Abstract: Large language models (LLMs) can solve arithmetic word problems with high accuracy, but little is known about how well they generalize to problems that are more complex than the ones on which they have been trained. Empirical investigations of such questions are impeded by two major flaws of current evaluations: (i) much of the evaluation data is contaminated, in the sense that it has already been seen during training, and (ii) benchmark datasets do not capture how problem proofs may be arbitrarily complex in various ways. As a step towards addressing these issues, we present a framework for evaluating LLMs on problems with arbitrarily complex arithmetic proofs, called MathGAP. MathGAP generates problems that follow fixed proof specifications -- along with chain-of-thought reasoning annotations -- enabling systematic studies on generalization with respect to arithmetic proof complexity. We apply MathGAP to analyze how in-context learning interacts with generalization to problems that have more complex proofs. We find that among the models tested, most show a significant decrease in performance as proofs get deeper and wider. This effect is more pronounced in complex, nonlinear proof structures, which are challenging even for GPT-4o. Surprisingly, providing in-context examples from the same distribution as the test set is not always beneficial for performance. In particular, zero-shot prompting as well as demonstrating a diverse range of examples that are less complex than the test data sometimes yield similar or higher accuracies.
Authors: Vinod Raman, Ambuj Tewari
Abstract: We study generation through the lens of statistical learning theory. First, we abstract and formalize the results of Gold [1967], Angluin [1979, 1980], and Kleinberg and Mullainathan [2024] for language identification/generation in the limit in terms of a binary hypothesis class defined over an abstract instance space. Then, we formalize a different paradigm of generation studied by Kleinberg and Mullainathan [2024], which we call ``uniform generation," and provide a characterization of which hypothesis classes are uniformly generatable. As is standard in statistical learning theory, our characterization is in terms of the finiteness of a new combinatorial dimension we call the Closure dimension. By doing so, we are able to compare generatability with predictability (captured via PAC and online learnability) and show that these two properties of hypothesis classes are \emph{incompatible} - there are classes that are generatable but not predictable and vice versa.
Authors: Argyrios Gerogiannis, Yu-Han Huang, Venugopal V. Veeravalli
Abstract: We study the problem of Non-Stationary Reinforcement Learning (NS-RL) without prior knowledge about the system's non-stationarity. A state-of-the-art, black-box algorithm, known as MASTER, is considered, with a focus on identifying the conditions under which it can achieve its stated goals. Specifically, we prove that MASTER's non-stationarity detection mechanism is not triggered for practical choices of horizon, leading to performance akin to a random restarting algorithm. Moreover, we show that the regret bound for MASTER, while being order optimal, stays above the worst-case linear regret until unreasonably large values of the horizon. To validate these observations, MASTER is tested for the special case of piecewise stationary multi-armed bandits, along with methods that employ random restarting, and others that use quickest change detection to restart. A simple, order optimal random restarting algorithm, that has prior knowledge of the non-stationarity is proposed as a baseline. The behavior of the MASTER algorithm is validated in simulations, and it is shown that methods employing quickest change detection are more robust and consistently outperform MASTER and other random restarting approaches.
Authors: Chen Bo Calvin Zhang, Zhang-Wei Hong, Aldo Pacchiano, Pulkit Agrawal
Abstract: Reward shaping is a critical component in reinforcement learning (RL), particularly for complex tasks where sparse rewards can hinder learning. While shaping rewards have been introduced to provide additional guidance, selecting effective shaping functions remains challenging and computationally expensive. This paper introduces Online Reward Selection and Policy Optimization (ORSO), a novel approach that frames shaping reward selection as an online model selection problem. ORSO employs principled exploration strategies to automatically identify promising shaping reward functions without human intervention, balancing exploration and exploitation with provable regret guarantees. We demonstrate ORSO's effectiveness across various continuous control tasks using the Isaac Gym simulator. Compared to traditional methods that fully evaluate each shaping reward function, ORSO significantly improves sample efficiency, reduces computational time, and consistently identifies high-quality reward functions that produce policies comparable to those generated by domain experts through hand-engineered rewards.
Authors: Yikang Chen, Dehui du, Lili Tian
Abstract: We propose an importance sampling method for tractable and efficient estimation of counterfactual expressions in general settings, named Exogenous Matching. By minimizing a common upper bound of counterfactual estimators, we transform the variance minimization problem into a conditional distribution learning problem, enabling its integration with existing conditional distribution modeling approaches. We validate the theoretical results through experiments under various types and settings of Structural Causal Models (SCMs) and demonstrate the outperformance on counterfactual estimation tasks compared to other existing importance sampling methods. We also explore the impact of injecting structural prior knowledge (counterfactual Markov boundaries) on the results. Finally, we apply this method to identifiable proxy SCMs and demonstrate the unbiasedness of the estimates, empirically illustrating the applicability of the method to practical scenarios.
Authors: Ethan Rathbun, Christopher Amato, Alina Oprea
Abstract: Recent works have demonstrated the vulnerability of Deep Reinforcement Learning (DRL) algorithms against training-time, backdoor poisoning attacks. These attacks induce pre-determined, adversarial behavior in the agent upon observing a fixed trigger during deployment while allowing the agent to solve its intended task during training. Prior attacks rely on arbitrarily large perturbations to the agent's rewards to achieve both of these objectives - leaving them open to detection. Thus, in this work, we propose a new class of backdoor attacks against DRL which achieve state of the art performance while minimally altering the agent's rewards. These "inception" attacks train the agent to associate the targeted adversarial behavior with high returns by inducing a disjunction between the agent's chosen action and the true action executed in the environment during training. We formally define these attacks and prove they can achieve both adversarial objectives. We then devise an online inception attack which significantly out-performs prior attacks under bounded reward constraints.
Authors: Mingqian Li, Qiao Han, Yiteng Zhai, Ruifeng Li, Yao Yang, Hongyang Chen
Abstract: In multi-task learning, we often encounter the case when the presence of labels across samples exhibits irregular patterns: samples can be fully labeled, partially labeled or unlabeled. Taking drug analysis as an example, multiple toxicity properties of a drug molecule may not be concurrently available due to experimental limitations. It triggers a demand for a new training and inference mechanism that could accommodate irregularly present labels and maximize the utility of any available label information. In this work, we focus on the two-label learning task, and propose a novel training and inference framework, Dual-Label Learning (DLL). The DLL framework formulates the problem into a dual-function system, in which the two functions should simultaneously satisfy standard supervision, structural duality and probabilistic duality. DLL features a dual-tower model architecture that explicitly captures the information exchange between labels, aimed at maximizing the utility of partially available labels in understanding label correlation. During training, label imputation for missing labels is conducted as part of the forward propagation process, while during inference, labels are regarded as unknowns of a bivariate system of equations and are solved jointly. Theoretical analysis guarantees the feasibility of DLL, and extensive experiments are conducted to verify that by explicitly modeling label correlation and maximizing the utility of available labels, our method makes consistently better predictions than baseline approaches by up to a 10% gain in F1-score or MAPE. Remarkably, our method provided with data at a label missing rate as high as 60% can achieve similar or even better results than baseline approaches at a label missing rate of only 10%.
Authors: Matthew J. Vowels, Mathieu Rochat, Sina Akbari
Abstract: Artificial Neural Networks (ANNs), including fully-connected networks and transformers, are highly flexible and powerful function approximators, widely applied in fields like computer vision and natural language processing. However, their inability to inherently respect causal structures can limit their robustness, making them vulnerable to covariate shift and difficult to interpret/explain. This poses significant challenges for their reliability in real-world applications. In this paper, we introduce Causal Fully-Connected Neural Networks (CFCNs) and Causal Transformers (CaTs), two general model families designed to operate under predefined causal constraints, as specified by a Directed Acyclic Graph (DAG). These models retain the powerful function approximation abilities of traditional neural networks while adhering to the underlying structural constraints, improving robustness, reliability, and interpretability at inference time. This approach opens new avenues for deploying neural networks in more demanding, real-world scenarios where robustness and explainability is critical.
Authors: Christina Lee Yu, Xumei Xi
Abstract: Tensor completion exhibits an interesting computational-statistical gap in terms of the number of samples needed to perform tensor estimation. While there are only $\Theta(tn)$ degrees of freedom in a $t$-order tensor with $n^t$ entries, the best known polynomial time algorithm requires $O(n^{t/2})$ samples in order to guarantee consistent estimation. In this paper, we show that weak side information is sufficient to reduce the sample complexity to $O(n)$. The side information consists of a weight vector for each of the modes which is not orthogonal to any of the latent factors along that mode; this is significantly weaker than assuming noisy knowledge of the subspaces. We provide an algorithm that utilizes this side information to produce a consistent estimator with $O(n^{1+\kappa})$ samples for any small constant $\kappa > 0$. We also provide experiments on both synthetic and real-world datasets that validate our theoretical insights.
Authors: Olcay Kursun, Ahmad Patooghy, Peyman Poursani, Oleg V. Favorov
Abstract: Contextually Guided Convolutional Neural Networks (CG-CNNs) employ self-supervision and contextual information to develop transferable features across diverse domains, including visual, tactile, temporal, and textual data. This work showcases the adaptability of CG-CNNs through applications to various datasets such as Caltech and Brodatz textures, the VibTac-12 tactile dataset, hyperspectral images, and challenges like the XOR problem and text analysis. In text analysis, CG-CNN employs an innovative embedding strategy that utilizes the context of neighboring words for classification, while in visual and signal data, it enhances feature extraction by exploiting spatial information. CG-CNN mimics the context-guided unsupervised learning mechanisms of biological neural networks and it can be trained to learn its features on limited-size datasets. Our experimental results on natural images reveal that CG-CNN outperforms comparable first-layer features of well-known deep networks such as AlexNet, ResNet, and GoogLeNet in terms of transferability and classification accuracy. In text analysis, CG-CNN learns word embeddings that outperform traditional models like Word2Vec in tasks such as the 20 Newsgroups text classification. Furthermore, ongoing development involves training CG-CNN on outputs from another CG-CNN to explore multi-layered architectures, aiming to construct more complex and descriptive features. This scalability and adaptability to various data types underscore the potential of CG-CNN to handle a wide range of applications, making it a promising architecture for tackling diverse data representation challenges.
Authors: Bj\"orn L\"utjens, Brandon Leshchinskiy, Oc\'eane Boulais, Farrukh Chishtie, Natalia D\'iaz-Rodr\'iguez, Margaux Masson-Forsythe, Ana Mata-Payerro, Christian Requena-Mesa, Aruna Sankaranarayanan, Aaron Pi\~na, Yarin Gal, Chedy Ra\"issi, Alexander Lavin, Dava Newman
Abstract: Deep generative vision models are now able to synthesize realistic-looking satellite imagery. But, the possibility of hallucinations prevents their adoption for risk-sensitive applications, such as generating materials for communicating climate change. To demonstrate this issue, we train a generative adversarial network (pix2pixHD) to create synthetic satellite imagery of future flooding and reforestation events. We find that a pure deep learning-based model can generate photorealistic flood visualizations but hallucinates floods at locations that were not susceptible to flooding. To address this issue, we propose to condition and evaluate generative vision models on segmentation maps of physics-based flood models. We show that our physics-conditioned model outperforms the pure deep learning-based model and a handcrafted baseline. We evaluate the generalization capability of our method to different remote sensing data and different climate-related events (reforestation). We publish our code and dataset which includes the data for a third case study of melting Arctic sea ice and $>$30,000 labeled HD image triplets -- or the equivalent of 5.5 million images at 128x128 pixels -- for segmentation guided image-to-image translation in Earth observation. Code and data is available at \url{https://github.com/blutjens/eie-earth-public}.
Authors: Mingzhang Yin, Yixin Wang, David M. Blei
Abstract: This paper presents a new optimization approach to causal estimation. Given data that contains covariates and an outcome, which covariates are causes of the outcome, and what is the strength of the causality? In classical machine learning (ML), the goal of optimization is to maximize predictive accuracy. However, some covariates might exhibit a non-causal association with the outcome. Such spurious associations provide predictive power for classical ML, but they prevent us from causally interpreting the result. This paper proposes CoCo, an optimization algorithm that bridges the gap between pure prediction and causal inference. CoCo leverages the recently-proposed idea of environments, datasets of covariates/response where the causal relationships remain invariant but where the distribution of the covariates changes from environment to environment. Given datasets from multiple environments-and ones that exhibit sufficient heterogeneity-CoCo maximizes an objective for which the only solution is the causal solution. We describe the theoretical foundations of this approach and demonstrate its effectiveness on simulated and real datasets. Compared to classical ML and existing methods, CoCo provides more accurate estimates of the causal model and more accurate predictions under interventions.
Authors: Wenxuan Wang, Wenxiang Jiao, Shuo Wang, Zhaopeng Tu, Michael R. Lyu
Abstract: Zero-shot translation is a promising direction for building a comprehensive multilingual neural machine translation~(MNMT) system. However, its quality is still not satisfactory due to off-target issues. In this paper, we aim to understand and alleviate the off-target issues from the perspective of uncertainty in zero-shot translation. By carefully examining the translation output and model confidence, we identify two uncertainties that are responsible for the off-target issues, namely, extrinsic data uncertainty and intrinsic model uncertainty. Based on the observations, we propose two lightweight and complementary approaches to denoise the training data for model training and explicitly penalize the off-target translations by unlikelihood training during model training. Extensive experiments on both balanced and imbalanced datasets show that our approaches significantly improve the performance of zero-shot translation over strong MNMT baselines.
Authors: Sheheryar Mehmood, Peter Ochs
Abstract: A large class of non-smooth practical optimization problems can be written as minimization of a sum of smooth and partly smooth functions. We examine such structured problems which also depend on a parameter vector and study the problem of differentiating its solution mapping with respect to the parameter which has far reaching applications in sensitivity analysis and parameter learning problems. Under partial smoothness and other mild assumptions, we apply Implicit (ID) and Automatic Differentiation (AD) to the fixed-point iterations of proximal splitting algorithms. We show that AD of the sequence generated by these algorithms converges (linearly under further assumptions) to the derivative of the solution mapping. For a variant of automatic differentiation, which we call Fixed-Point Automatic Differentiation (FPAD), we remedy the memory overhead problem of the Reverse Mode AD and moreover provide faster convergence theoretically. We numerically illustrate the convergence and convergence rates of AD and FPAD on Lasso and Group Lasso problems and demonstrate the working of FPAD on prototypical image denoising problems by learning the regularization term.
Authors: Dmitrijs Trizna
Abstract: We propose a hybrid machine learning architecture that simultaneously employs multiple deep learning models analyzing contextual and behavioral characteristics of Windows portable executable, producing a final prediction based on a decision from the meta-model. The detection heuristic in contemporary machine learning Windows malware classifiers is typically based on the static properties of the sample since dynamic analysis through virtualization is challenging for vast quantities of samples. To surpass this limitation, we employ a Windows kernel emulation that allows the acquisition of behavioral patterns across large corpora with minimal temporal and computational costs. We partner with a security vendor for a collection of more than 100k int-the-wild samples that resemble the contemporary threat landscape, containing raw PE files and filepaths of applications at the moment of execution. The acquired dataset is at least ten folds larger than reported in related works on behavioral malware analysis. Files in the training dataset are labeled by a professional threat intelligence team, utilizing manual and automated reverse engineering tools. We estimate the hybrid classifier's operational utility by collecting an out-of-sample test set three months later from the acquisition of the training set. We report an improved detection rate, above the capabilities of the current state-of-the-art model, especially under low false-positive requirements. Additionally, we uncover a meta-model's ability to identify malicious activity in validation and test sets even if none of the individual models express enough confidence to mark the sample as malevolent. We conclude that the meta-model can learn patterns typical to malicious samples from representation combinations produced by different analysis techniques. We publicly release pre-trained models and anonymized dataset of emulation reports.
Authors: Chandrajit Bajaj, Minh Nguyen, Conrad Li
Abstract: In this paper, we present a novel reinforcement learning framework designed to optimize molecular dynamics by focusing on the entire trajectory rather than just the final molecular configuration. Leveraging a stochastic version of Pontryagin's Maximum Principle (PMP) and Soft Actor-Critic (SAC) algorithm, our framework effectively explores non-convex molecular energy landscapes, escaping local minima to stabilize in low-energy states. Our approach operates in continuous state and action spaces without relying on labeled data, making it applicable to a wide range of molecular systems. Through extensive experimentation on six distinct molecules, including Bradykinin and Oxytocin, we demonstrate competitive performance against other unsupervised physics-based methods, such as the Greedy and NEMO-based algorithms. Our method's adaptability and focus on dynamic trajectory optimization make it suitable for applications in areas such as drug discovery and molecular design.
Authors: Yuen Chen, Vethavikashini Chithrra Raghuram, Justus Mattern, Rada Mihalcea, Zhijing Jin
Abstract: Generated texts from large language models (LLMs) have been shown to exhibit a variety of harmful, human-like biases against various demographics. These findings motivate research efforts aiming to understand and measure such effects. This paper introduces a causal formulation for bias measurement in generative language models. Based on this theoretical foundation, we outline a list of desiderata for designing robust bias benchmarks. We then propose a benchmark called OccuGender, with a bias-measuring procedure to investigate occupational gender bias. We test several state-of-the-art open-source LLMs on OccuGender, including Llama, Mistral, and their instruction-tuned versions. The results show that these models exhibit substantial occupational gender bias. Lastly, we discuss prompting strategies for bias mitigation and an extension of our causal formulation to illustrate the generalizability of our framework. Our code and data https://github.com/chenyuen0103/gender-bias.
Authors: Jason M. Altschuler, Kunal Talwar
Abstract: A canonical algorithm for log-concave sampling is the Langevin Algorithm, aka the Langevin Diffusion run with some discretization stepsize $\eta > 0$. This discretization leads the Langevin Algorithm to have a stationary distribution $\pi_{\eta}$ which differs from the stationary distribution $\pi$ of the Langevin Diffusion, and it is an important challenge to understand whether the well-known properties of $\pi$ extend to $\pi_{\eta}$. In particular, while concentration properties such as isoperimetry and rapidly decaying tails are classically known for $\pi$, the analogous properties for $\pi_{\eta}$ are open questions with algorithmic implications. This note provides a first step in this direction by establishing concentration results for $\pi_{\eta}$ that mirror classical results for $\pi$. Specifically, we show that for any nontrivial stepsize $\eta > 0$, $\pi_{\eta}$ is sub-exponential (respectively, sub-Gaussian) when the potential is convex (respectively, strongly convex). Moreover, the concentration bounds we show are essentially tight. We also show that these concentration bounds extend to all iterates along the trajectory of the Langevin Algorithm, and to inexact implementations which use sub-Gaussian estimates of the gradient. Key to our analysis is the use of a rotation-invariant moment generating function (aka Bessel function) to study the stationary dynamics of the Langevin Algorithm. This technique may be of independent interest because it enables directly analyzing the discrete-time stationary distribution $\pi_{\eta}$ without going through the continuous-time stationary distribution $\pi$ as an intermediary.
Authors: Martin Andersson, Benny Avelin
Abstract: We develop theory and methods that use the graph Laplacian to analyze the geometry of the underlying manifold of datasets. Our theory provides theoretical guarantees and explicit bounds on the functional forms of the graph Laplacian when it acts on functions defined close to singularities of the underlying manifold. We use these explicit bounds to develop tests for singularities and propose methods that can be used to estimate geometric properties of singularities in the datasets.
Authors: Gang Qiao, Ambuj Tewari
Abstract: We study the convex hull membership (CHM) problem in the pure exploration setting where one aims to efficiently and accurately determine if a given point lies in the convex hull of means of a finite set of distributions. We give a complete characterization of the sample complexity of the CHM problem in the one-dimensional case. We present the first asymptotically optimal algorithm called Thompson-CHM, whose modular design consists of a stopping rule and a sampling rule. In addition, we extend the algorithm to settings that generalize several important problems in the multi-armed bandit literature. Furthermore, we discuss the extension of Thompson-CHM to higher dimensions. Finally, we provide numerical experiments to demonstrate the empirical behavior of the algorithm matches our theoretical results for realistic time horizons.
Authors: Noboru Isobe, Mizuho Okumura
Abstract: This paper presents a mathematical analysis of ODE-Net, a continuum model of deep neural networks (DNNs). In recent years, Machine Learning researchers have introduced ideas of replacing the deep structure of DNNs with ODEs as a continuum limit. These studies regard the "learning" of ODE-Net as the minimization of a "loss" constrained by a parametric ODE. Although the existence of a minimizer for this minimization problem needs to be assumed, only a few studies have investigated its existence analytically in detail. In the present paper, the existence of a minimizer is discussed based on a formulation of ODE-Net as a measure-theoretic mean-field optimal control problem. The existence result is proved when a neural network, which describes a vector field of ODE-Net, is linear with respect to learnable parameters. The proof employs the measure-theoretic formulation combined with the direct method of Calculus of Variations. Secondly, an idealized minimization problem is proposed to remove the above linearity assumption. Such a problem is inspired by a kinetic regularization associated with the Benamou--Brenier formula and universal approximation theorems for neural networks. The proofs of these existence results use variational methods, differential equations, and mean-field optimal control theory. They will stand for a new analytic way to investigate the learning process of deep neural networks.
Authors: Jonas Ngnawe, Marianne Abemgnigni Njifon, Jonathan Heek, Yann Dauphin
Abstract: Deep networks have achieved impressive results on a range of well-curated benchmark datasets. Surprisingly, their performance remains sensitive to perturbations that have little effect on human performance. In this work, we propose a novel extension of Mixup called Robustmix that regularizes networks to classify based on lower-frequency spatial features. We show that this type of regularization improves robustness on a range of benchmarks such as Imagenet-C and Stylized Imagenet. It adds little computational overhead and, furthermore, does not require a priori knowledge of a large set of image transformations. We find that this approach further complements recent advances in model architecture and data augmentation, attaining a state-of-the-art mCE of 44.8 with an EfficientNet-B8 model and RandAugment, which is a reduction of 16 mCE compared to the baseline.
Authors: Vincent Zhihao Zheng, Seongjin Choi, Lijun Sun
Abstract: Deep probabilistic time series forecasting has gained attention for its ability to provide nonlinear approximation and valuable uncertainty quantification for decision-making. However, existing models often oversimplify the problem by assuming a time-independent error process and overlooking serial correlation. To overcome this limitation, we propose an innovative training method that incorporates error autocorrelation to enhance probabilistic forecasting accuracy. Our method constructs a mini-batch as a collection of $D$ consecutive time series segments for model training. It explicitly learns a time-varying covariance matrix over each mini-batch, encoding error correlation among adjacent time steps. The learned covariance matrix can be used to improve prediction accuracy and enhance uncertainty quantification. We evaluate our method on two different neural forecasting models and multiple public datasets. Experimental results confirm the effectiveness of the proposed approach in improving the performance of both models across a range of datasets, resulting in notable improvements in predictive accuracy.
Authors: Xitong Zhang, Avrajit Ghosh, Guangliang Liu, Rongrong Wang
Abstract: Previous research on PAC-Bayes learning theory has focused extensively on establishing tight upper bounds for test errors. A recently proposed training procedure called PAC-Bayes training, updates the model toward minimizing these bounds. Although this approach is theoretically sound, in practice, it has not achieved a test error as low as those obtained by empirical risk minimization (ERM) with carefully tuned regularization hyperparameters. Additionally, existing PAC-Bayes training algorithms often require bounded loss functions and may need a search over priors with additional datasets, which limits their broader applicability. In this paper, we introduce a new PAC-Bayes training algorithm with improved performance and reduced reliance on prior tuning. This is achieved by establishing a new PAC-Bayes bound for unbounded loss and a theoretically grounded approach that involves jointly training the prior and posterior using the same dataset. Our comprehensive evaluations across various classification tasks and neural network architectures demonstrate that the proposed method not only outperforms existing PAC-Bayes training algorithms but also approximately matches the test accuracy of ERM that is optimized by SGD/Adam using various regularization methods with optimal hyperparameters.
Authors: Marcin Przewi\k{e}\'zlikowski, Mateusz Pyla, Bartosz Zieli\'nski, Bart{\l}omiej Twardowski, Jacek Tabor, Marek \'Smieja
Abstract: Self-supervised learning (SSL) is a powerful technique for learning from unlabeled data. By learning to remain invariant to applied data augmentations, methods such as SimCLR and MoCo can reach quality on par with supervised approaches. However, this invariance may be detrimental for solving downstream tasks that depend on traits affected by augmentations used during pretraining, such as color. In this paper, we propose to foster sensitivity to such characteristics in the representation space by modifying the projector network, a common component of self-supervised architectures. Specifically, we supplement the projector with information about augmentations applied to images. For the projector to take advantage of this auxiliary conditioning when solving the SSL task, the feature extractor learns to preserve the augmentation information in its representations. Our approach, coined Conditional Augmentation-aware Self-supervised Learning (CASSLE), is directly applicable to typical joint-embedding SSL methods regardless of their objective functions. Moreover, it does not require major changes in the network architecture or prior knowledge of downstream tasks. In addition to an analysis of sensitivity towards different data augmentations, we conduct a series of experiments, which show that CASSLE improves over various SSL methods, reaching state-of-the-art performance in multiple downstream tasks.
Authors: Xiao Guo, Xiang Li, Xiangyu Chang, Shujie Ma
Abstract: Modern multi-layer networks are commonly stored and analyzed in a local and distributed fashion because of the privacy, ownership, and communication costs. The literature on the model-based statistical methods for community detection based on these data is still limited. This paper proposes a new method for consensus community detection and estimation in a multi-layer stochastic block model using locally stored and computed network data with privacy protection. A novel algorithm named privacy-preserving Distributed Spectral Clustering (ppDSC) is developed. To preserve the edges' privacy, we adopt the randomized response (RR) mechanism to perturb the network edges, which satisfies the strong notion of differential privacy. The ppDSC algorithm is performed on the squared RR-perturbed adjacency matrices to prevent possible cancellation of communities among different layers. To remove the bias incurred by RR and the squared network matrices, we develop a two-step bias-adjustment procedure. Then we perform eigen-decomposition on the debiased matrices, aggregation of the local eigenvectors using an orthogonal Procrustes transformation, and k-means clustering. We provide theoretical analysis on the statistical errors of ppDSC in terms of eigen-vector estimation. In addition, the blessings and curses of network heterogeneity are well-explained by our bounds.
Authors: Anav Sood, Trevor Hastie
Abstract: We consider the problem of selecting a small subset of representative variables from a large dataset. In the computer science literature, this dimensionality reduction problem is typically formalized as Column Subset Selection (CSS). Meanwhile, the typical statistical formalization is to find an information-maximizing set of Principal Variables. This paper shows that these two approaches are equivalent, and moreover, both can be viewed as maximum likelihood estimation within a certain semi-parametric model. Within this model, we establish suitable conditions under which the CSS estimate is consistent in high dimensions, specifically in the proportional asymptotic regime where the number of variables over the sample size converges to a constant. Using these connections, we show how to efficiently (1) perform CSS using only summary statistics from the original dataset; (2) perform CSS in the presence of missing and/or censored data; and (3) select the subset size for CSS in a hypothesis testing framework.
Authors: Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu, Mubarak Shah
Abstract: Text-to-image diffusion models have recently attracted the interest of many researchers, and inverting the diffusion process can play an important role in better understanding the generative process and how to engineer prompts in order to obtain the desired images. To this end, we study the task of predicting the prompt embedding given an image generated by a generative diffusion model. We consider a series of white-box and black-box models (with and without access to the weights of the diffusion network) to deal with the proposed task. We propose a novel learning framework comprising a joint prompt regression and multi-label vocabulary classification objective that generates improved prompts. To further improve our method, we employ a curriculum learning procedure that promotes the learning of image-prompt pairs with lower labeling noise (i.e. that are better aligned). We conduct experiments on the DiffusionDB data set, predicting text prompts from images generated by Stable Diffusion. In addition, we make an interesting discovery: training a diffusion model on the prompt generation task can make the model generate images that are much better aligned with the input prompts, when the model is directly reused for text-to-image generation. Our code is publicly available for download at https://github.com/CroitoruAlin/Reverse-Stable-Diffusion.
URLs: https://github.com/CroitoruAlin/Reverse-Stable-Diffusion.
Authors: Ayan Sengupta, Md Shad Akhtar, Tanmoy Chakraborty
Abstract: Code-mixing and script-mixing are prevalent across online social networks and multilingual societies. However, a user's preference toward code-mixing depends on the socioeconomic status, demographics of the user, and the local context, which existing generative models mostly ignore while generating code-mixed texts. In this work, we make a pioneering attempt to develop a persona-aware generative model to generate texts resembling real-life code-mixed texts of individuals. We propose a Persona-aware Generative Model for Code-mixed Generation, PARADOX, a novel Transformer-based encoder-decoder model that encodes an utterance conditioned on a user's persona and generates code-mixed texts without monolingual reference data. We propose an alignment module that re-calibrates the generated sequence to resemble real-life code-mixed texts. PARADOX generates code-mixed texts that are semantically more meaningful and linguistically more valid. To evaluate the personification capabilities of PARADOX, we propose four new metrics -- CM BLEU, CM Rouge-1, CM Rouge-L and CM KS. On average, PARADOX achieves 1.6 points better CM BLEU, 47% better perplexity and 32% better semantic coherence than the non-persona-based counterparts.
Authors: Kai Gao, Zhaxizhuoma, Yan Ding, Shiqi Zhang, Jingjin Yu
Abstract: Effectively performing object rearrangement is an essential skill for mobile manipulators, e.g., setting up a dinner table or organizing a desk. A key challenge in such problems is deciding an appropriate manipulation order for objects to effectively untangle dependencies between objects while considering the necessary motions for realizing the manipulations (e.g., pick and place). To our knowledge, computing time-optimal multi-object rearrangement solutions for mobile manipulators remains a largely untapped research direction. In this research, we propose ORLA*, which leverages delayed (lazy) evaluation in searching for a high-quality object pick and place sequence that considers both end-effector and mobile robot base travel. ORLA* also supports multi-layered rearrangement tasks considering pile stability using machine learning. Employing an optimal solver for finding temporary locations for displacing objects, ORLA* can achieve global optimality. Through extensive simulation and ablation study, we confirm the effectiveness of ORLA* delivering quality solutions for challenging rearrangement instances. Supplementary materials are available at: https://gaokai15.github.io/ORLA-Star/
Authors: Tomoki Nishi, Yusuke Hara
Abstract: Discrete-choice models are a powerful framework for analyzing decision-making behavior to provide valuable insights for policymakers and businesses. Multinomial logit models (MNLs) with linear utility functions have been used in practice because they are ease to use and interpretable. Recently, MNLs with neural networks (e.g., ASU-DNN) have been developed, and they have achieved higher prediction accuracy in behavior choice than classical MNLs. However, these models lack interpretability owing to complex structures. We developed utility functions with a novel neural-network architecture based on generalized additive models, named generalized additive utility network ( GAUNet), for discrete-choice models. We evaluated the performance of the MNL with GAUNet using the trip survey data collected in Tokyo. Our models were comparable to ASU-DNN in accuracy and exhibited improved interpretability compared to previous models.
Authors: Jingkang Yang, Yuhao Dong, Shuai Liu, Bo Li, Ziyue Wang, Chencheng Jiang, Haoran Tan, Jiamu Kang, Yuanhan Zhang, Kaiyang Zhou, Ziwei Liu
Abstract: Large vision-language models (VLMs) have achieved substantial progress in multimodal perception and reasoning. When integrated into an embodied agent, existing embodied VLM works either output detailed action sequences at the manipulation level or only provide plans at an abstract level, leaving a gap between high-level planning and real-world manipulation. To bridge this gap, we introduce Octopus, an embodied vision-language programmer that uses executable code generation as a medium to connect planning and manipulation. Octopus is designed to 1) proficiently comprehend an agent's visual and textual task objectives, 2) formulate intricate action sequences, and 3) generate executable code. To facilitate Octopus model development, we introduce OctoVerse: a suite of environments tailored for benchmarking vision-based code generators on a wide spectrum of tasks, ranging from mundane daily chores in simulators to sophisticated interactions in complex video games such as Grand Theft Auto (GTA) and Minecraft. To train Octopus, we leverage GPT-4 to control an explorative agent that generates training data, i.e., action blueprints and corresponding executable code. We also collect feedback that enables an enhanced training scheme called Reinforcement Learning with Environmental Feedback (RLEF). Through a series of experiments, we demonstrate Octopus's functionality and present compelling results, showing that the proposed RLEF refines the agent's decision-making. By open-sourcing our simulation environments, dataset, and model architecture, we aspire to ignite further innovation and foster collaborative applications within the broader embodied AI community.
Authors: James Halverson, Fabian Ruehle
Abstract: We develop a general theory of flows in the space of Riemannian metrics induced by neural network gradient descent. This is motivated in part by recent advances in approximating Calabi-Yau metrics with neural networks and is enabled by recent advances in understanding flows in the space of neural networks. We derive the corresponding metric flow equations, which are governed by a metric neural tangent kernel, a complicated, non-local object that evolves in time. However, many architectures admit an infinite-width limit in which the kernel becomes fixed and the dynamics simplify. Additional assumptions can induce locality in the flow, which allows for the realization of Perelman's formulation of Ricci flow that was used to resolve the 3d Poincar\'e conjecture. We demonstrate that such fixed kernel regimes lead to poor learning of numerical Calabi-Yau metrics, as is expected since the associated neural networks do not learn features. Conversely, we demonstrate that well-learned numerical metrics at finite-width exhibit an evolving metric-NTK, associated with feature learning. Our theory of neural network metric flows therefore explains why neural networks are better at learning Calabi-Yau metrics than fixed kernel methods, such as the Ricci flow.
Authors: Yalin E. Sagduyu, Tugba Erpek, Aylin Yener, Sennur Ulukus
Abstract: This paper explores the integration of deep learning techniques for joint sensing and communications, with an extension to semantic communications. The integrated system comprises a transmitter and receiver operating over a wireless channel, subject to noise and fading. The transmitter employs a deep neural network (DNN), namely an encoder, for joint operations of source coding, channel coding, and modulation, while the receiver utilizes another DNN, namely a decoder, for joint operations of demodulation, channel decoding, and source decoding to reconstruct the data samples. The transmitted signal serves a dual purpose, supporting communication with the receiver and enabling sensing. When a target is present, the reflected signal is received, and another DNN decoder is utilized for sensing. This decoder is responsible for detecting the target's presence and determining its range. All these DNNs, including one encoder and two decoders, undergo joint training through multi-task learning, considering data and channel characteristics. This paper extends to incorporate semantic communications by introducing an additional DNN, another decoder at the receiver, operating as a task classifier. This decoder evaluates the fidelity of label classification for received signals, enhancing the integration of semantics within the communication process. The study presents results based on using the CIFAR-10 as the input data and accounting for channel effects like Additive White Gaussian Noise (AWGN) and Rayleigh fading. The results underscore the effectiveness of multi-task deep learning in achieving high-fidelity joint sensing and semantic communications.
Authors: Shadab Ahamed, Yixi Xu, Claire Gowdy, Joo H. O, Ingrid Bloise, Don Wilson, Patrick Martineau, Fran\c{c}ois B\'enard, Fereshteh Yousefirizi, Rahul Dodhia, Juan M. Lavista, William B. Weeks, Carlos F. Uribe, Arman Rahmim
Abstract: This study performs comprehensive evaluation of four neural network architectures (UNet, SegResNet, DynUNet, and SwinUNETR) for lymphoma lesion segmentation from PET/CT images. These networks were trained, validated, and tested on a diverse, multi-institutional dataset of 611 cases. Internal testing (88 cases; total metabolic tumor volume (TMTV) range [0.52, 2300] ml) showed SegResNet as the top performer with a median Dice similarity coefficient (DSC) of 0.76 and median false positive volume (FPV) of 4.55 ml; all networks had a median false negative volume (FNV) of 0 ml. On the unseen external test set (145 cases with TMTV range: [0.10, 2480] ml), SegResNet achieved the best median DSC of 0.68 and FPV of 21.46 ml, while UNet had the best FNV of 0.41 ml. We assessed reproducibility of six lesion measures, calculated their prediction errors, and examined DSC performance in relation to these lesion measures, offering insights into segmentation accuracy and clinical relevance. Additionally, we introduced three lesion detection criteria, addressing the clinical need for identifying lesions, counting them, and segmenting based on metabolic characteristics. We also performed expert intra-observer variability analysis revealing the challenges in segmenting ``easy'' vs. ``hard'' cases, to assist in the development of more resilient segmentation algorithms. Finally, we performed inter-observer agreement assessment underscoring the importance of a standardized ground truth segmentation protocol involving multiple expert annotators. Code is available at: https://github.com/microsoft/lymphoma-segmentation-dnn
URLs: https://github.com/microsoft/lymphoma-segmentation-dnn
Authors: Anirudh Ajith, Sameer Singh, Danish Pruthi
Abstract: Watermarking involves implanting an imperceptible signal into generated text that can later be detected via statistical tests. A prominent family of watermarking strategies for LLMs embeds this signal by upsampling a (pseudorandomly-chosen) subset of tokens at every generation step. However, such signals alter the model's output distribution and can have unintended effects on its downstream performance. In this work, we evaluate the performance of LLMs watermarked using three different strategies over a diverse suite of tasks including those cast as k-class classification (CLS), multiple choice question answering (MCQ), short-form generation (e.g., open-ended question answering) and long-form generation (e.g., translation) tasks. We find that watermarks (under realistic hyperparameters) can cause significant drops in LLMs' effective utility across all tasks. We observe drops of 10 to 20% in CLS tasks in the average case, which shoot up to 100% in the worst case. We notice degradations of about 7% in MCQ tasks, 10-15% in short-form generation, and 5-15% in long-form generation tasks. Our findings highlight the trade-offs that users should be cognizant of when using watermarked models.
Authors: William T. Redman, Dean Huang, Maria Fonoberova, Igor Mezi\'c
Abstract: Koopman operator theory has found significant success in learning models of complex, real-world dynamical systems, enabling prediction and control. The greater interpretability and lower computational costs of these models, compared to traditional machine learning methodologies, make Koopman learning an especially appealing approach. Despite this, little work has been performed on endowing Koopman learning with the ability to leverage its own failures. To address this, we equip Koopman methods - developed for predicting non-autonomous time-series - with an episodic memory mechanism, enabling global recall of (or attention to) periods in time where similar dynamics previously occurred. We find that a basic implementation of Koopman learning with episodic memory leads to significant improvements in prediction on synthetic and real-world data. Our framework has considerable potential for expansion, allowing for future advances, and opens exciting new directions for Koopman learning.
Authors: Marlon Tobaben, Hibiki Ito, Joonas J\"alk\"o, Gauri Pradhan, Yuan He, Antti Honkela
Abstract: We analyse the relationship between privacy vulnerability and dataset properties, such as examples per class and number of classes, when applying two state-of-the-art membership inference attacks (MIAs) to fine-tuned neural networks. We derive per-example MIA vulnerability in terms of score distributions and statistics computed from shadow models. We introduce a simplified model of membership inference and prove that in this model, the logarithm of the difference of true and false positive rates depends linearly on the logarithm of the number of examples per class. We complement the theoretical analysis with empirical analysis by systematically testing the practical privacy vulnerability of fine-tuning large image classification models and obtain the previously derived power law dependence between the number of examples per class in the data and the MIA vulnerability, as measured by true positive rate of the attack at a low false positive rate. Finally, we fit a parametric model of the previously derived form to predict true positive rate based on dataset properties and observe good fit for MIA vulnerability on unseen fine-tuning scenarios.
Authors: Joshua C Chang, Xiangting Li, Shixin Xu, Hao-Ren Yao, Julia Porcino, Carson Chow
Abstract: We introduce gradient-flow-guided adaptive importance sampling (IS) transformations for stabilizing Monte-Carlo approximations of leave-one-out (LOO) cross-validated predictions for Bayesian models. After defining two variational problems, we derive corresponding simple nonlinear transformations that utilize gradient information to shift a model's pre-trained full-data posterior closer to the target LOO posterior predictive distributions. In doing so, the transformations stabilize importance weights. The resulting Monte Carlo integrals depend on Jacobian determinants with respect to the model Hessian. We derive closed-form exact formulae for these Jacobian determinants in the cases of logistic regression and shallow ReLU-activated artificial neural networks, and provide a simple approximation that sidesteps the need to compute full Hessian matrices and their spectra. We test the methodology on an $n\ll p$ dataset that is known to produce unstable LOO IS weights.
Authors: Pei-Hung Chung, Shuhan He, Norawit Kijpaisalratana, Abdel-badih el Ariss, Byung-Jun Yoon
Abstract: A Clinical Decision Support System (CDSS) is designed to enhance clinician decision-making by combining system-generated recommendations with medical expertise. Given the high costs, intensive labor, and time-sensitive nature of medical treatments, there is a pressing need for efficient decision support, especially in complex emergency scenarios. In these scenarios, where information can be limited, an advanced CDSS framework that leverages AI (artificial intelligence) models to effectively reduce diagnostic uncertainty has utility. Such an AI-enabled CDSS framework with quantified uncertainty promises to be practical and beneficial in the demanding context of real-world medical care. In this study, we introduce the concept of Medical Entropy, quantifying uncertainties in patient outcomes predicted by neural machine translation based on the ICD-9 code of procedures. Our experimental results not only show strong correlations between procedure and diagnosis sequences based on the simple ICD-9 code but also demonstrate the promising capacity to model trends of uncertainties during hospitalizations through a data-driven approach.
Authors: Matan Avitan, Ryan Cotterell, Yoav Goldberg, Shauli Ravfogel
Abstract: Interventions targeting the representation space of language models (LMs) have emerged as an effective means to influence model behavior. Such methods are employed, for example, to eliminate or alter the encoding of demographic information such as gender within the model's representations and, in so doing, create a counterfactual representation. However, because the intervention operates within the representation space, understanding precisely what aspects of the text it modifies poses a challenge. In this paper, we give a method to convert representation counterfactuals into string counterfactuals. We demonstrate that this approach enables us to analyze the linguistic alterations corresponding to a given representation space intervention and to interpret the features utilized to encode a specific concept. Moreover, the resulting counterfactuals can be used to mitigate bias in classification through data augmentation.
Authors: Yunwei Bai, Ying Kiat Tan, Shiming Chen, Yao Shu, Tsuhan Chen
Abstract: Few-shot-learning (FSL) commonly requires a model to identify images (queries) that belong to classes unseen during training, based on a few labeled samples of the new classes (support set) as reference. So far, plenty of algorithms involve training data augmentation to improve the generalization capability of FSL models, but outlier queries or support images during inference can still pose great generalization challenges. In this work, to reduce the bias caused by the outlier samples, we generate additional test-class samples by combining original samples with suitable train-class samples via a generative image combiner. Then, we obtain averaged features via an augmentor, which leads to more typical representations through the averaging. We experimentally and theoretically demonstrate the effectiveness of our method, e.g., obtaining a test accuracy improvement proportion of around 10% (e.g., from 46.86% to 53.28%) for trained FSL models. Importantly, given pretrained image combiner, our method is training-free for off-the-shelf FSL models, whose performance can be improved without extra datasets nor further training of the models themselves.
Authors: Marta Campi, Guillaume Staerman, Gareth W. Peters, Tomoko Matsui
Abstract: Functional Isolation Forest (FIF) is a recent state-of-the-art Anomaly Detection (AD) algorithm designed for functional data. It relies on a tree partition procedure where an abnormality score is computed by projecting each curve observation on a drawn dictionary through a linear inner product. Such linear inner product and the dictionary are a priori choices that highly influence the algorithm's performances and might lead to unreliable results, particularly with complex datasets. This work addresses these challenges by introducing \textit{Signature Isolation Forest}, a novel AD algorithm class leveraging the rough path theory's signature transform. Our objective is to remove the constraints imposed by FIF through the proposition of two algorithms which specifically target the linearity of the FIF inner product and the choice of the dictionary. We provide several numerical experiments, including a real-world applications benchmark showing the relevance of our methods.
Authors: Jan Pe\v{s}\'an, Santosh Kesiraju, Luk\'a\v{s} Burget, Jan ''Honza'' \v{C}ernock\'y
Abstract: Paralinguistic traits like cognitive load and emotion are increasingly recognized as pivotal areas in speech recognition research, often examined through specialized datasets like CLSE and IEMOCAP. However, the integrity of these datasets is seldom scrutinized for text-dependency. This paper critically evaluates the prevalent assumption that machine learning models trained on such datasets genuinely learn to identify paralinguistic traits, rather than merely capturing lexical features. By examining the lexical overlap in these datasets and testing the performance of machine learning models, we expose significant text-dependency in trait-labeling. Our results suggest that some machine learning models, especially large pre-trained models like HuBERT, might inadvertently focus on lexical characteristics rather than the intended paralinguistic features. The study serves as a call to action for the research community to reevaluate the reliability of existing datasets and methodologies, ensuring that machine learning models genuinely learn what they are designed to recognize.
Authors: Anthony Liang, Jesse Thomason, Erdem B{\i}y{\i}k
Abstract: Training robots to perform complex control tasks from high-dimensional pixel input using reinforcement learning (RL) is sample-inefficient, because image observations are comprised primarily of task-irrelevant information. By contrast, humans are able to visually attend to task-relevant objects and areas. Based on this insight, we introduce Visual Saliency-Guided Reinforcement Learning (ViSaRL). Using ViSaRL to learn visual representations significantly improves the success rate, sample efficiency, and generalization of an RL agent on diverse tasks including DeepMind Control benchmark, robot manipulation in simulation and on a real robot. We present approaches for incorporating saliency into both CNN and Transformer-based encoders. We show that visual representations learned using ViSaRL are robust to various sources of visual perturbations including perceptual noise and scene variations. ViSaRL nearly doubles success rate on the real-robot tasks compared to the baseline which does not use saliency.
Authors: Ziyu Wang, Chris Holmes
Abstract: Bayesian modelling allows for the quantification of predictive uncertainty which is crucial in safety-critical applications. Yet for many machine learning (ML) algorithms, it is difficult to construct or implement their Bayesian counterpart. In this work we present a promising approach to address this challenge, based on the hypothesis that commonly used ML algorithms are efficient across a wide variety of tasks and may thus be near Bayes-optimal w.r.t. an unknown task distribution. We prove that it is possible to recover the Bayesian posterior defined by the task distribution, which is unknown but optimal in this setting, by building a martingale posterior using the algorithm. We further propose a practical uncertainty quantification method that apply to general ML algorithms. Experiments based on a variety of non-NN and NN algorithms demonstrate the efficacy of our method.
Authors: Chuyi Shang, Amos You, Sanjay Subramanian, Trevor Darrell, Roei Herzig
Abstract: Recently, image-based Large Multimodal Models (LMMs) have made significant progress in video question-answering (VideoQA) using a frame-wise approach by leveraging large-scale pretraining in a zero-shot manner. Nevertheless, these models need to be capable of finding relevant information, extracting it, and answering the question simultaneously. Currently, existing methods perform all of these steps in a single pass without being able to adapt if insufficient or incorrect information is collected. To overcome this, we introduce a modular multi-LMM agent framework based on several agents with different roles, instructed by a Planner agent that updates its instructions using shared feedback from the other agents. Specifically, we propose TraveLER, a method that can create a plan to "Traverse" through the video, ask questions about individual frames to "Locate" and store key information, and then "Evaluate" if there is enough information to answer the question. Finally, if there is not enough information, our method is able to "Replan" based on its collected knowledge. Through extensive experiments, we find that the proposed TraveLER approach improves performance on several VideoQA benchmarks without the need to fine-tune on specific datasets. Our code is available at https://github.com/traveler-framework/TraveLER.
Authors: Chin-Yun Yu, Christopher Mitcheltree, Alistair Carson, Stefan Bilbao, Joshua D. Reiss, Gy\"orgy Fazekas
Abstract: Infinite impulse response filters are an essential building block of many time-varying audio systems, such as audio effects and synthesisers. However, their recursive structure impedes end-to-end training of these systems using automatic differentiation. Although non-recursive filter approximations like frequency sampling and frame-based processing have been proposed and widely used in previous works, they cannot accurately reflect the gradient of the original system. We alleviate this difficulty by re-expressing a time-varying all-pole filter to backpropagate the gradients through itself, so the filter implementation is not bound to the technical limitations of automatic differentiation frameworks. This implementation can be employed within audio systems containing filters with poles for efficient gradient evaluation. We demonstrate its training efficiency and expressive capabilities for modelling real-world dynamic audio systems on a phaser, time-varying subtractive synthesiser, and compressor. We make our code and audio samples available and provide the trained audio effect and synth models in a VST plugin at https://diffapf.github.io/web/.
Authors: Yiwen Tang, Ray Zhang, Jiaming Liu, Zoey Guo, Dong Wang, Zhigang Wang, Bin Zhao, Shanghang Zhang, Peng Gao, Hongsheng Li, Xuelong Li
Abstract: Large foundation models have recently emerged as a prominent focus of interest, attaining superior performance in widespread scenarios. Due to the scarcity of 3D data, many efforts have been made to adapt pre-trained transformers from vision to 3D domains. However, such 2D-to-3D approaches are still limited, due to the potential loss of spatial geometries and high computation cost. More importantly, their frameworks are mainly designed for 2D models, lacking a general any-to-3D paradigm. In this paper, we introduce Any2Point, a parameter-efficient method to empower any-modality large models (vision, language, audio) for 3D understanding. Given a frozen transformer from any source modality, we propose a 3D-to-any (1D or 2D) virtual projection strategy that correlates the input 3D points to the original 1D or 2D positions within the source modality. This mechanism enables us to assign each 3D token with a positional encoding paired with the pre-trained model, which avoids 3D geometry loss caused by the true projection and better motivates the transformer for 3D learning with 1D/2D positional priors. Then, within each transformer block, we insert an any-to-3D guided adapter module for parameter-efficient fine-tuning. The adapter incorporates prior spatial knowledge from the source modality to guide the local feature aggregation of 3D tokens, compelling the semantic adaption of any-modality transformers. We conduct extensive experiments to showcase the effectiveness and efficiency of our method. Code and models are released at https://github.com/Ivan-Tang-3D/Any2Point.
Authors: Shirley Wu, Shiyu Zhao, Michihiro Yasunaga, Kexin Huang, Kaidi Cao, Qian Huang, Vassilis N. Ioannidis, Karthik Subbian, James Zou, Jure Leskovec
Abstract: Answering real-world complex queries, such as complex product search, often requires accurate retrieval from semi-structured knowledge bases that involve blend of unstructured (e.g., textual descriptions of products) and structured (e.g., entity relations of products) information. However, many previous works studied textual and relational retrieval tasks as separate topics. To address the gap, we develop STARK, a large-scale Semi-structure retrieval benchmark on Textual and Relational Knowledge Bases. Our benchmark covers three domains: product search, academic paper search, and queries in precision medicine. We design a novel pipeline to synthesize realistic user queries that integrate diverse relational information and complex textual properties, together with their ground-truth answers (items). We conduct rigorous human evaluation to validate the quality of our synthesized queries. We further enhance the benchmark with high-quality human-generated queries to provide an authentic reference. STARK serves as a comprehensive testbed for evaluating the performance of retrieval systems driven by large language models (LLMs). Our experiments suggest that STARK presents significant challenges to the current retrieval and LLM systems, highlighting the need for more capable semi-structured retrieval systems. The benchmark data and code are available on https://github.com/snap-stanford/STaRK.
Authors: Mauricio Lima, Katherine Deck, Oliver R. A. Dunbar, Tapio Schneider
Abstract: Machine learning is playing an increasing role in hydrology, supplementing or replacing physics-based models. One notable example is the use of recurrent neural networks (RNNs) for forecasting streamflow given observed precipitation and geographic characteristics. Training of such a model over the continental United States (CONUS) demonstrated that a single set of model parameters can be used across independent catchments, and that RNNs can outperform physics-based models. In this work, we take a next step and study the performance of RNNs for river routing in land surface models (LSMs). Instead of observed precipitation, the LSM-RNN uses instantaneous runoff calculated from physics-based models as an input. We train the model with data from river basins spanning the globe and test using historical streamflow measurements. The model demonstrates skill at generalization across basins (predicting streamflow in catchments not used in training) and across time (predicting streamflow during years not used in training). We compare the predictions from the LSM-RNN to an existing physics-based model calibrated with a similar dataset and find that the LSM-RNN outperforms the physics-based model. Our results show that RNNs are effective for global streamflow prediction from runoff inputs and motivate the development of complete routing models that can capture nested sub-basis connections.
Authors: Juyong Lee, Taywon Min, Minyong An, Dongyoon Hahm, Haeone Lee, Changyeon Kim, Kimin Lee
Abstract: Mobile device control agents can largely enhance user interactions and productivity by automating daily tasks. However, despite growing interest in developing practical agents, the absence of a commonly adopted benchmark in this area makes it challenging to quantify scientific progress. In this work, we introduce B-MoCA: a novel benchmark with interactive environments for evaluating and developing mobile device control agents. To create a realistic benchmark, we develop B-MoCA based on the Android operating system and define 131 common daily tasks. Importantly, we incorporate a randomization feature that changes the configurations of mobile devices, including user interface layouts and language settings, to assess generalization performance. We benchmark diverse agents, including agents employing large language models (LLMs) or multi-modal LLMs as well as agents trained with imitation learning using human expert demonstrations. While these agents demonstrate proficiency in executing straightforward tasks, their poor performance on complex tasks highlights significant opportunities for future research to improve effectiveness. Our source code is publicly available at https://b-moca.github.io.
Authors: Zhiyu Guo, Hidetaka Kamigaito, Taro Wanatnabe
Abstract: The rapid advancement in Large Language Models (LLMs) has markedly enhanced the capabilities of language understanding and generation. However, the substantial model size poses hardware challenges, affecting both memory size for serving and inference latency for token generation. To address those challenges, we propose Dependency-aware Semi-structured Sparsity (DaSS), a novel method for the recent prevalent GLU-based LLMs pruning, which incorporates structural dependency into the weight magnitude-based unstructured pruning. We introduce an MLP-specific pruning metric that evaluates the importance of each weight by jointly considering its magnitude and its corresponding MLP intermediate activation norms. DaSS facilitates a balance between the adaptability offered by unstructured pruning and the structural consistency inherent in dependency-based structured pruning. Empirical evaluations on LLaMA2, Mistral, and Gemma model families demonstrate that DaSS not only outperforms both SparseGPT and Wanda in achieving hardware-friendly N:M sparsity patterns but also maintains the computational efficiency of Wanda.
Authors: Inkee Jung, Siu-Cheong Lau
Abstract: In this paper,we develop a local-to-global and measure-theoretical approach to understand datasets. The idea is to take network models with restricted domains as local charts of datasets. We develop the mathematical foundations for these structures, and show in experiments how it can be used to find fuzzy domains and to improve accuracy in data classification problems.
Authors: Ruchit Rawal, Khalid Saifullah, Miquel Farr\'e, Ronen Basri, David Jacobs, Gowthami Somepalli, Tom Goldstein
Abstract: Current datasets for long-form video understanding often fall short of providing genuine long-form comprehension challenges, as many tasks derived from these datasets can be successfully tackled by analyzing just one or a few random frames from a video. To address this issue, we present a novel dataset and benchmark, CinePile, specifically designed for authentic long-form video understanding. This paper details our innovative approach for creating a question-answer dataset, utilizing advanced LLMs with human-in-the-loop and building upon human-generated raw data. Our comprehensive dataset comprises 305,000 multiple-choice questions (MCQs), covering various visual and multimodal aspects, including temporal comprehension, understanding human-object interactions, and reasoning about events or actions within a scene. Additionally, we fine-tuned open-source Video-LLMs on the training split and evaluated both open-source and proprietary video-centric LLMs on the test split of our dataset. The findings indicate that although current models underperform compared to humans, fine-tuning these models can lead to significant improvements in their performance.
Authors: Anthony Bardou, Patrick Thiran, Giovanni Ranieri
Abstract: Bayesian Optimization (BO) has proven to be very successful at optimizing a static, noisy, costly-to-evaluate black-box function $f : \mathcal{S} \to \mathbb{R}$. However, optimizing a black-box which is also a function of time (i.e., a dynamic function) $f : \mathcal{S} \times \mathcal{T} \to \mathbb{R}$ remains a challenge, since a dynamic Bayesian Optimization (DBO) algorithm has to keep track of the optimum over time. This changes the nature of the optimization problem in at least three aspects: (i) querying an arbitrary point in $\mathcal{S} \times \mathcal{T}$ is impossible, (ii) past observations become less and less relevant for keeping track of the optimum as time goes by and (iii) the DBO algorithm must have a high sampling frequency so it can collect enough relevant observations to keep track of the optimum through time. In this paper, we design a Wasserstein distance-based criterion able to quantify the relevancy of an observation with respect to future predictions. Then, we leverage this criterion to build W-DBO, a DBO algorithm able to remove irrelevant observations from its dataset on the fly, thus maintaining simultaneously a good predictive performance and a high sampling frequency, even in continuous-time optimization tasks with unknown horizon. Numerical experiments establish the superiority of W-DBO, which outperforms state-of-the-art methods by a comfortable margin.
Authors: Amin Sarihi, Peter Jamieson, Ahmad Patooghy, Abdel-Hameed A. Badawy
Abstract: The Hardware Trojan (HT) problem can be thought of as a continuous game between attackers and defenders, each striving to outsmart the other by leveraging any available means for an advantage. Machine Learning (ML) has recently played a key role in advancing HT research. Various novel techniques, such as Reinforcement Learning (RL) and Graph Neural Networks (GNNs), have shown HT insertion and detection capabilities. HT insertion with ML techniques, specifically, has seen a spike in research activity due to the shortcomings of conventional HT benchmarks and the inherent human design bias that occurs when we create them. This work continues this innovation by presenting a tool called TrojanForge, capable of generating HT adversarial examples that defeat HT detectors; demonstrating the capabilities of GAN-like adversarial tools for automatic HT insertion. We introduce an RL environment where the RL insertion agent interacts with HT detectors in an insertion-detection loop where the agent collects rewards based on its success in bypassing HT detectors. Our results show that this process helps inserted HTs evade various HT detectors, achieving high attack success percentages. This tool provides insight into why HT insertion fails in some instances and how we can leverage this knowledge in defense.
Authors: Isha Chaudhary, Qian Hu, Manoj Kumar, Morteza Ziyadi, Rahul Gupta, Gagandeep Singh
Abstract: Large Language Models (LLMs) can produce biased responses that can cause representational harms. However, conventional studies are insufficient to thoroughly evaluate LLM bias, as they can not scale to large number of inputs and provide no guarantees. Therefore, we propose the first framework, QuaCer-B that certifies LLMs for bias on distributions of prompts. A certificate consists of high-confidence bounds on the probability of unbiased LLM responses for any set of prompts mentioning various demographic groups, sampled from a distribution. We illustrate the bias certification for distributions of prompts created by applying varying prefixes drawn from a prefix distributions, to a given set of prompts. We consider prefix distributions for random token sequences, mixtures of manual jailbreaks, and jailbreaks in the LLM's embedding space to certify bias. We obtain non-trivial certified bounds on the probability of unbiased responses of SOTA LLMs, exposing their vulnerabilities over distributions of prompts generated from computationally inexpensive distributions of prefixes.
Authors: Mingmeng Geng, Sihong He, Roberto Trotta
Abstract: Can large language models (LLMs) simulate social surveys? To answer this question, we conducted millions of simulations in which LLMs were asked to answer subjective questions. A comparison of different LLM responses with the European Social Survey (ESS) data suggests that the effect of prompts on bias and variability is fundamental, highlighting major cultural, age, and gender biases. We further discussed statistical methods for measuring the difference between LLM answers and survey data and proposed a novel measure inspired by Jaccard similarity, as LLM-generated responses are likely to have a smaller variance. Our experiments also reveal that it is important to analyze the robustness and variability of prompts before using LLMs to simulate social surveys, as their imitation abilities are approximate at best.
Authors: Yihe Wang, Nan Huang, Taida Li, Yujun Yan, Xiang Zhang
Abstract: Medical time series (MedTS) data, such as Electroencephalography (EEG) and Electrocardiography (ECG), play a crucial role in healthcare, such as diagnosing brain and heart diseases. Existing methods for MedTS classification primarily rely on handcrafted biomarkers extraction and CNN-based models, with limited exploration of transformer-based models. In this paper, we introduce Medformer, a multi-granularity patching transformer tailored specifically for MedTS classification. Our method incorporates three novel mechanisms to leverage the unique characteristics of MedTS: cross-channel patching to leverage inter-channel correlations, multi-granularity embedding for capturing features at different scales, and two-stage (intra- and inter-granularity) multi-granularity self-attention for learning features and correlations within and among granularities. We conduct extensive experiments on five public datasets under both subject-dependent and challenging subject-independent setups. Results demonstrate Medformer's superiority over 10 baselines, achieving top averaged ranking across five datasets on all six evaluation metrics. These findings underscore the significant impact of our method on healthcare applications, such as diagnosing Myocardial Infarction, Alzheimer's, and Parkinson's disease. We release the source code at https://github.com/DL4mHealth/Medformer.
Authors: Richard Luo, Austin Peng, Adithya Vasudev, Rishabh Jain
Abstract: Video is an increasingly prominent and information-dense medium, yet it poses substantial challenges for language models. A typical video consists of a sequence of shorter segments, or shots, that collectively form a coherent narrative. Each shot is analogous to a word in a sentence where multiple data streams of information (such as visual and auditory data) must be processed simultaneously. Comprehension of the entire video requires not only understanding the visual-audio information of each shot but also requires that the model links the ideas between each shot to generate a larger, all-encompassing story. Despite significant progress in the field, current works often overlook videos' more granular shot-by-shot semantic information. In this project, we propose a family of efficient large language vision models (LLVMs) to boost video summarization and captioning called Shotluck Holmes. By leveraging better pretraining and data collection strategies, we extend the abilities of existing small LLVMs from being able to understand a picture to being able to understand a sequence of frames. Specifically, we show that Shotluck Holmes achieves better performance than state-of-the-art results on the Shot2Story video captioning and summary task with significantly smaller and more computationally efficient models.
Authors: Seyed Moein Ayyoubzadeh, Kourosh Shahnazari
Abstract: Homograph disambiguation, the task of distinguishing words with identical spellings but different meanings, poses a substantial challenge in natural language processing. In this study, we introduce a novel dataset tailored for Persian homograph disambiguation. Our work encompasses a thorough exploration of various embeddings, evaluated through the cosine similarity method and their efficacy in downstream tasks like classification. Our investigation entails training a diverse array of lightweight machine learning and deep learning models for phonograph disambiguation. We scrutinize the models' performance in terms of Accuracy, Recall, and F1 Score, thereby gaining insights into their respective strengths and limitations. The outcomes of our research underscore three key contributions. First, we present a newly curated Persian dataset, providing a solid foundation for future research in homograph disambiguation. Second, our comparative analysis of embeddings highlights their utility in different contexts, enriching the understanding of their capabilities. Third, by training and evaluating a spectrum of models, we extend valuable guidance for practitioners in selecting suitable strategies for homograph disambiguation tasks. In summary, our study unveils a new dataset, scrutinizes embeddings through diverse perspectives, and benchmarks various models for homograph disambiguation. These findings empower researchers and practitioners to navigate the intricate landscape of homograph-related challenges effectively.
Authors: Sriram Balasubramanian, Samyadeep Basu, Soheil Feizi
Abstract: Recent work has explored how individual components of the CLIP-ViT model contribute to the final representation by leveraging the shared image-text representation space of CLIP. These components, such as attention heads and MLPs, have been shown to capture distinct image features like shape, color or texture. However, understanding the role of these components in arbitrary vision transformers (ViTs) is challenging. To this end, we introduce a general framework which can identify the roles of various components in ViTs beyond CLIP. Specifically, we (a) automate the decomposition of the final representation into contributions from different model components, and (b) linearly map these contributions to CLIP space to interpret them via text. Additionally, we introduce a novel scoring function to rank components by their importance with respect to specific features. Applying our framework to various ViT variants (e.g. DeiT, DINO, DINOv2, Swin, MaxViT), we gain insights into the roles of different components concerning particular image features. These insights facilitate applications such as image retrieval using text descriptions or reference images, visualizing token importance heatmaps, and mitigating spurious correlations. We release our code to reproduce the experiments at https://github.com/SriramB-98/vit-decompose
Authors: Chengyuan Deng, Yiqun Duan, Xin Jin, Heng Chang, Yijun Tian, Han Liu, Yichen Wang, Kuofeng Gao, Henry Peng Zou, Yiqiao Jin, Yijia Xiao, Shenghao Wu, Zongxing Xie, Weimin Lyu, Sihong He, Lu Cheng, Haohan Wang, Jun Zhuang
Abstract: Large Language Models (LLMs) have achieved unparalleled success across diverse language modeling tasks in recent years. However, this progress has also intensified ethical concerns, impacting the deployment of LLMs in everyday contexts. This paper provides a comprehensive survey of ethical challenges associated with LLMs, from longstanding issues such as copyright infringement, systematic bias, and data privacy, to emerging problems like truthfulness and social norms. We critically analyze existing research aimed at understanding, examining, and mitigating these ethical risks. Our survey underscores integrating ethical standards and societal values into the development of LLMs, thereby guiding the development of responsible and ethically aligned language models.
Authors: Felix O'Mahony, Yulong Yang, Christine Allen-Blanchette
Abstract: In this paper, we introduce group convolutional neural networks (GCNNs) equivariant to color variation. GCNNs have been designed for a variety of geometric transformations from 2D and 3D rotation groups, to semi-groups such as scale. Despite the improved interpretability, accuracy and generalizability of these architectures, GCNNs have seen limited application in the context of perceptual quantities. Notably, the recent CEConv network uses a GCNN to achieve equivariance to hue transformations by convolving input images with a hue rotated RGB filter. However, this approach leads to invalid RGB values which break equivariance and degrade performance. We resolve these issues with a lifting layer that transforms the input image directly, thereby circumventing the issue of invalid RGB values and improving equivariance error by over three orders of magnitude. Moreover, we extend the notion of color equivariance to include equivariance to saturation shift. Our hue-, saturation-, and color-equivariant networks achieve strong generalization to out-of-distribution perceptual variations and improved sample efficiency over conventional architectures. We demonstrate the utility of our approach on synthetic and real world datasets where we consistently outperform competitive baselines.
Authors: Soroush Saghafian, Lihi Idan
Abstract: Advanced analytics science methods have enabled combining the power of artificial and human intelligence, creating \textit{centaurs} that allow superior decision-making. Centaurs are hybrid human-algorithm models that combine both formal analytics and human intuition in a symbiotic manner within their learning and reasoning process. We argue that the future of AI development and use in many domains needs to focus more on centaurs as opposed to other AI approaches. This paradigm shift towards centaur-based AI methods raises some fundamental questions: How are centaurs different from other human-in-the-loop methods? What are the most effective methods for creating centaurs? When should centaurs be used, and when should the lead be given to pure AI models? Doesn't the incorporation of human intuition -- which at times can be misleading -- in centaurs' decision-making process degrade its performance compared to pure AI methods? This work aims to address these fundamental questions, focusing on recent advancements in generative AI, and especially in Large Language Models (LLMs), as a main case study to illustrate centaurs' critical essentiality to future AI endeavors.
Authors: Amelie Gyrard, Seyedali Mohammadi, Manas Gaur, Antonio Kung
Abstract: Sustainable Development Goals (SDGs) give the UN a road map for development with Agenda 2030 as a target. SDG3 "Good Health and Well-Being" ensures healthy lives and promotes well-being for all ages. Digital technologies can support SDG3. Burnout and even depression could be reduced by encouraging better preventive health. Due to the lack of patient knowledge and focus to take care of their health, it is necessary to help patients before it is too late. New trends such as positive psychology and mindfulness are highly encouraged in the USA. Digital Twins (DTs) can help with the continuous monitoring of emotion using physiological signals (e.g., collected via wearables). DTs facilitate monitoring and provide constant health insight to improve quality of life and well-being with better personalization. Healthcare DTs challenges are standardizing data formats, communication protocols, and data exchange mechanisms. As an example, ISO has the ISO/IEC JTC 1/SC 41 Internet of Things (IoT) and DTs Working Group, with standards such as "ISO/IEC 21823-3:2021 IoT - Interoperability for IoT Systems - Part 3 Semantic interoperability", "ISO/IEC CD 30178 - IoT - Data format, value and coding". To achieve those data integration and knowledge challenges, we designed the Mental Health Knowledge Graph (ontology and dataset) to boost mental health. As an example, explicit knowledge is described such as chocolate contains magnesium which is recommended for depression. The Knowledge Graph (KG) acquires knowledge from ontology-based mental health projects classified within the LOV4IoT ontology catalog (Emotion, Depression, and Mental Health). Furthermore, the KG is mapped to standards when possible. Standards from ETSI SmartM2M can be used such as SAREF4EHAW to represent medical devices and sensors, but also ITU/WHO, ISO, W3C, NIST, and IEEE standards relevant to mental health can be considered.
Authors: Michail Chatzianastasis, Yang Zhang, George Dasoulas, Michalis Vazirgiannis
Abstract: Protein representation learning aims to learn informative protein embeddings capable of addressing crucial biological questions, such as protein function prediction. Although sequence-based transformer models have shown promising results by leveraging the vast amount of protein sequence data in a self-supervised way, there is still a gap in exploiting the available 3D protein structures. In this work, we propose a pre-training scheme going beyond trivial masking methods leveraging 3D and hierarchical structures of proteins. We propose a novel self-supervised method to pretrain 3D graph neural networks on 3D protein structures, by predicting the distances between local geometric centroids of protein subgraphs and the global geometric centroid of the protein. By considering subgraphs and their relationships to the global protein structure, our model can better learn the geometric properties of the protein structure. We experimentally show that our proposed pertaining strategy leads to significant improvements up to 6\%, in the performance of 3D GNNs in various protein classification tasks. Our work opens new possibilities in unsupervised learning for protein graph models while eliminating the need for multiple views, augmentations, or masking strategies which are currently used so far.
Authors: Sungjun Lim, Jeyoon Yeom, Sooyon Kim, Hoyoon Byun, Jinho Kang, Yohan Jung, Jiyoung Jung, Kyungwoo Song
Abstract: Bayesian neural network (BNN) approximates the posterior distribution of model parameters and utilizes the posterior for prediction via Bayesian Model Averaging (BMA). The quality of the posterior approximation is critical for achieving accurate and robust predictions. It is known that flatness in the loss landscape is strongly associated with generalization performance, and it necessitates consideration to improve the quality of the posterior approximation. In this work, we empirically demonstrate that BNNs often struggle to capture the flatness. Moreover, we provide both experimental and theoretical evidence showing that BMA can be ineffective without ensuring flatness. To address this, we propose Sharpness-Aware Bayesian Model Averaging (SA-BMA), a novel optimizer that seeks flat posteriors by calculating divergence in the parameter space. SA-BMA aligns with the intrinsic nature of BNN and the generalized version of existing sharpness-aware optimizers for DNN. In addition, we suggest a Bayesian Transfer Learning scheme to efficiently leverage pre-trained DNN. We validate the efficacy of SA-BMA in enhancing generalization performance in few-shot classification and distribution shift by ensuring flat posterior.
Authors: Trung Trinh, Markus Heinonen, Luigi Acerbi, Samuel Kaski
Abstract: Deep neural networks (DNNs) excel on clean images but struggle with corrupted ones. Incorporating specific corruptions into the data augmentation pipeline can improve robustness to those corruptions but may harm performance on clean images and other types of distortion. In this paper, we introduce an alternative approach that improves the robustness of DNNs to a wide range of corruptions without compromising accuracy on clean images. We first demonstrate that input perturbations can be mimicked by multiplicative perturbations in the weight space. Leveraging this, we propose Data Augmentation via Multiplicative Perturbation (DAMP), a training method that optimizes DNNs under random multiplicative weight perturbations. We also examine the recently proposed Adaptive Sharpness-Aware Minimization (ASAM) and show that it optimizes DNNs under adversarial multiplicative weight perturbations. Experiments on image classification datasets (CIFAR-10/100, TinyImageNet and ImageNet) and neural network architectures (ResNet50, ViT-S/16, ViT-B/16) show that DAMP enhances model generalization performance in the presence of corruptions across different settings. Notably, DAMP is able to train a ViT-S/16 on ImageNet from scratch, reaching the top-1 error of 23.7% which is comparable to ResNet50 without extensive data augmentations.
Authors: Ka Man Lo, Zeyu Huang, Zihan Qiu, Zili Wang, Jie Fu
Abstract: Mixture-of-experts (MoE) is gaining increasing attention due to its unique properties and remarkable performance, especially for language tasks. By sparsely activating a subset of parameters for each token, MoE architecture could increase the model size without sacrificing computational efficiency, achieving a better trade-off between performance and training costs. However, the underlying mechanism of MoE still lacks further exploration, and its modularization degree remains questionable. In this paper, we make an initial attempt to understand the inner workings of MoE-based large language models. Concretely, we comprehensively study the parametric and behavioral features of three popular MoE-based models and reveal some intriguing observations, including 1) Neurons act like fine-grained experts; 2) The router of MoE usually selects experts with larger output norms; 3) The expert diversity increases as the layer increases, while the last layer is an outlier, which is further validated by an initial experiment. Based on the observations, we also provide suggestions for a broad spectrum of MoE practitioners, such as router design and expert allocation. We hope this work could shed light on future research on the MoE framework and other modular architectures. Code is available at https://github.com/kamanphoebe/Look-into-MoEs.
Authors: Shengcheng Luo, Quanquan Peng, Jun Lv, Kaiwen Hong, Katherine Rose Driggs-Campbell, Cewu Lu, Yong-Lu Li
Abstract: Employing a teleoperation system for gathering demonstrations offers the potential for more efficient learning of robot manipulation. However, teleoperating a robot arm equipped with a dexterous hand or gripper, via a teleoperation system presents inherent challenges due to the task's high dimensionality, complexity of motion, and differences between physiological structures. In this study, we introduce a novel system for joint learning between human operators and robots, that enables human operators to share control of a robot end-effector with a learned assistive agent, simplifies the data collection process, and facilitates simultaneous human demonstration collection and robot manipulation training. As data accumulates, the assistive agent gradually learns. Consequently, less human effort and attention are required, enhancing the efficiency of the data collection process. It also allows the human operator to adjust the control ratio to achieve a trade-off between manual and automated control. We conducted experiments in both simulated environments and physical real-world settings. Through user studies and quantitative evaluations, it is evident that the proposed system could enhance data collection efficiency and reduce the need for human adaptation while ensuring the collected data is of sufficient quality for downstream tasks. \textit{For more details, please refer to our webpage https://norweig1an.github.io/HAJL.github.io/.
Authors: Zhiyuan Wang, Jinhao Duan, Lu Cheng, Yue Zhang, Qingni Wang, Xiaoshuang Shi, Kaidi Xu, Hengtao Shen, Xiaofeng Zhu
Abstract: Uncertainty quantification (UQ) in natural language generation (NLG) tasks remains an open challenge, exacerbated by the closed-source nature of the latest large language models (LLMs). This study investigates applying conformal prediction (CP), which can transform any heuristic uncertainty notion into rigorous prediction sets, to black-box LLMs in open-ended NLG tasks. We introduce a novel uncertainty measure based on self-consistency theory, and then develop a conformal uncertainty criterion by integrating the uncertainty condition aligned with correctness into the CP algorithm. Empirical evaluations indicate that our uncertainty measure outperforms prior state-of-the-art methods. Furthermore, we achieve strict control over the correctness coverage rate utilizing 7 popular LLMs on 4 free-form NLG datasets, spanning general-purpose and medical scenarios. Additionally, the calibrated prediction sets with small size further highlights the efficiency of our method in providing trustworthy guarantees for practical open-ended NLG applications.
Authors: Lorenzo Bonasera, Emilio Carrizosa
Abstract: Tree ensemble methods represent a popular machine learning model, known for their effectiveness in supervised classification and regression tasks. Their performance derives from aggregating predictions of multiple decision trees, which are renowned for their interpretability properties. However, tree ensemble methods do not reliably exhibit interpretable output. Our work aims to extract an optimized list of rules from a trained tree ensemble, providing the user with a condensed, interpretable model that retains most of the predictive power of the full model. Our approach consists of solving a clean and neat set partitioning problem formulated through Integer Programming. The proposed method works with either tabular or time series data, for both classification and regression tasks, and does not require parameter tuning under the most common setting. Through rigorous computational experiments, we offer statistically significant evidence that our method is competitive with other rule extraction methods and effectively handles time series.
Authors: Nathan Cloos, Moufan Li, Markus Siegel, Scott L. Brincat, Earl K. Miller, Guangyu Robert Yang, Christopher J. Cueva
Abstract: How do we know if two systems - biological or artificial - process information in a similar way? Similarity measures such as linear regression, Centered Kernel Alignment (CKA), Normalized Bures Similarity (NBS), and angular Procrustes distance, are often used to quantify this similarity. However, it is currently unclear what drives high similarity scores and even what constitutes a "good" score. Here, we introduce a novel tool to investigate these questions by differentiating through similarity measures to directly maximize the score. Surprisingly, we find that high similarity scores do not guarantee encoding task-relevant information in a manner consistent with neural data; and this is particularly acute for CKA and even some variations of cross-validated and regularized linear regression. We find no consistent threshold for a good similarity score - it depends on both the measure and the dataset. In addition, synthetic datasets optimized to maximize similarity scores initially learn the highest variance principal component of the target dataset, but some methods like angular Procrustes capture lower variance dimensions much earlier than methods like CKA. To shed light on this, we mathematically derive the sensitivity of CKA, angular Procrustes, and NBS to the variance of principal component dimensions, and explain the emphasis CKA places on high variance components. Finally, by jointly optimizing multiple similarity measures, we characterize their allowable ranges and reveal that some similarity measures are more constraining than others. While current measures offer a seemingly straightforward way to quantify the similarity between neural systems, our work underscores the need for careful interpretation. We hope the tools we developed will be used by practitioners to better understand current and future similarity measures.
Authors: Zeliang Zhang, Xiaodong Liu, Hao Cheng, Chenliang Xu, Jianfeng Gao
Abstract: By increasing model parameters but activating them sparsely when performing a task, the use of Mixture-of-Experts (MoE) architecture significantly improves the performance of Large Language Models (LLMs) without increasing the inference cost. However, the memory consumption due to the growing number of experts presents a challenge to the deployment of these models in many real world settings. Our empirical study reveals that some experts encode redundant knowledge during pre-training. We thus propose a method of grouping and pruning similar experts to improve the model's parameter efficiency. We validate the effectiveness of our method by pruning three state-of-the-art MoE architectures, including Mixtral, Deepseek-MoE, and Qwen. The evaluation shows that our method outperforms other model pruning methods on a range of natural language tasks. We will release our code to facilitate future research.
Authors: Jan Ole von Hartz, Tim Welschehold, Abhinav Valada, Joschka Boedecker
Abstract: Task Parametrized Gaussian Mixture Models (TP-GMM) are a sample-efficient method for learning object-centric robot manipulation tasks. However, there are several open challenges to applying TP-GMMs in the wild. In this work, we tackle three crucial challenges synergistically. First, end-effector velocities are non-Euclidean and thus hard to model using standard GMMs. We thus propose to factorize the robot's end-effector velocity into its direction and magnitude, and model them using Riemannian GMMs. Second, we leverage the factorized velocities to segment and sequence skills from complex demonstration trajectories. Through the segmentation, we further align skill trajectories and hence leverage time as a powerful inductive bias. Third, we present a method to automatically detect relevant task parameters per skill from visual observations. Our approach enables learning complex manipulation tasks from just five demonstrations while using only RGB-D observations. Extensive experimental evaluations on RLBench demonstrate that our approach achieves state-of-the-art performance with 20-fold improved sample efficiency. Our policies generalize across different environments, object instances, and object positions, while the learned skills are reusable.
Authors: Amir Mohammad Karimi Mamaghan, Samuele Papa, Karl Henrik Johansson, Stefan Bauer, Andrea Dittadi
Abstract: Object-centric (OC) representations, which represent the state of a visual scene by modeling it as a composition of objects, have the potential to be used in various downstream tasks to achieve systematic compositional generalization and facilitate reasoning. However, these claims have not been thoroughly analyzed yet. Recently, foundation models have demonstrated unparalleled capabilities across diverse domains from language to computer vision, marking them as a potential cornerstone of future research for a multitude of computational tasks. In this paper, we conduct an extensive empirical study on representation learning for downstream Visual Question Answering (VQA), which requires an accurate compositional understanding of the scene. We thoroughly investigate the benefits and trade-offs of OC models and alternative approaches including large pre-trained foundation models on both synthetic and real-world data, and demonstrate a viable way to achieve the best of both worlds. The extensiveness of our study, encompassing over 600 downstream VQA models and 15 different types of upstream representations, also provides several additional insights that we believe will be of interest to the community at large.
Authors: Esteban Garces Arias, Julian Rodemann, Meimingwei Li, Christian Heumann, Matthias A{\ss}enmacher
Abstract: Decoding from the output distributions of large language models to produce high-quality text is a complex challenge in language modeling. Various approaches, such as beam search, sampling with temperature, $k-$sampling, nucleus $p-$sampling, typical decoding, contrastive decoding, and contrastive search, have been proposed to address this problem, aiming to improve coherence, diversity, as well as resemblance to human-generated text. In this study, we introduce adaptive contrastive search, a novel decoding strategy extending contrastive search by incorporating an adaptive degeneration penalty, guided by the estimated uncertainty of the model at each generation step. This strategy is designed to enhance both the creativity and diversity of the language modeling process while at the same time producing coherent and high-quality generated text output. Our findings indicate performance enhancement in both aspects, across different model architectures and datasets, underscoring the effectiveness of our method in text generation tasks. Our code base, datasets, and models are publicly available.
Authors: Nicholas Konz, Maciej A. Mazurowski
Abstract: In recent years, there has been interest in how geometric properties such as intrinsic dimension (ID) of a neural network's hidden representations change through its layers, and how such properties are predictive of important model behavior such as generalization ability. However, evidence has begun to emerge that such behavior can change significantly depending on the domain of the network's training data, such as natural versus medical images. Here, we further this inquiry by exploring how the ID of a network's learned representations changes through its layers, in essence, characterizing how the network successively refines the information content of input data to be used for predictions. Analyzing eleven natural and medical image datasets across six network architectures, we find that how ID changes through the network differs noticeably between natural and medical image models. Specifically, medical image models peak in representation ID earlier in the network, implying a difference in the image features and their abstractness that are typically used for downstream tasks in these domains. Additionally, we discover a strong correlation of this peak representation ID with the ID of the data in its input space, implying that the intrinsic information content of a model's learned representations is guided by that of the data it was trained on. Overall, our findings emphasize notable discrepancies in network behavior between natural and non-natural imaging domains regarding hidden representation information content, and provide further insights into how a network's learned features are shaped by its training data.
Authors: Yuejiang Liu, Jubayer Ibn Hamid, Annie Xie, Yoonho Lee, Maximilian Du, Chelsea Finn
Abstract: Predicting and executing a sequence of actions without intermediate replanning, known as action chunking, is increasingly used in robot learning from human demonstrations. Yet, its reported effects on the learned policy are inconsistent: some studies find it crucial for achieving strong results, while others observe decreased performance. In this paper, we first dissect how action chunking impacts the divergence between a learner and a demonstrator. We find that action chunking allows the learner to better capture the temporal dependencies in demonstrations but at the cost of reduced reactivity in stochastic environments. To address this tradeoff, we propose Bidirectional Decoding (BID), a test-time inference algorithm that bridges action chunking with closed-loop operations. BID samples multiple predictions at each time step and searches for the optimal one based on two criteria: (i) backward coherence, which favors samples that align with previous decisions; (ii) forward contrast, which seeks samples of high likelihood for future plans. By coupling decisions within and across action chunks, BID promotes consistency over time while maintaining reactivity to unexpected changes. Experimental results show that BID boosts the performance of two state-of-the-art generative policies across seven simulation benchmarks and two real-world tasks. Code and videos are available at https://bid-robot.github.io.
Authors: Lutfi Eren Erdogan, Nicholas Lee, Siddharth Jha, Sehoon Kim, Ryan Tabrizi, Suhong Moon, Coleman Hooper, Gopala Anumanchipalli, Kurt Keutzer, Amir Gholami
Abstract: Recent large language models (LLMs) have enabled the development of advanced agentic systems that can integrate various tools and APIs to fulfill user queries through function calling. However, the deployment of these LLMs on the edge has not been explored since they typically require cloud-based infrastructure due to their substantial model size and computational demands. To this end, we present TinyAgent, an end-to-end framework for training and deploying task-specific small language model agents capable of function calling for driving agentic systems at the edge. We first show how to enable accurate function calling for open-source models via the LLMCompiler framework. We then systematically curate a high-quality dataset for function calling, which we use to fine-tune two small language models, TinyAgent-1.1B and 7B. For efficient inference, we introduce a novel tool retrieval method to reduce the input prompt length and utilize quantization to further accelerate the inference speed. As a driving application, we demonstrate a local Siri-like system for Apple's MacBook that can execute user commands through text or voice input. Our results show that our models can achieve, and even surpass, the function-calling capabilities of larger models like GPT-4-Turbo, while being fully deployed at the edge. We open-source our dataset, models, and installable package and provide a demo video for our MacBook assistant agent.
Authors: Yuancheng Wang, Haoyue Zhan, Liwei Liu, Ruihong Zeng, Haotian Guo, Jiachen Zheng, Qiang Zhang, Xueyao Zhang, Shunsi Zhang, Zhizheng Wu
Abstract: The recent large-scale text-to-speech (TTS) systems are usually grouped as autoregressive and non-autoregressive systems. The autoregressive systems implicitly model duration but exhibit certain deficiencies in robustness and lack of duration controllability. Non-autoregressive systems require explicit alignment information between text and speech during training and predict durations for linguistic units (e.g. phone), which may compromise their naturalness. In this paper, we introduce Masked Generative Codec Transformer (MaskGCT), a fully non-autoregressive TTS model that eliminates the need for explicit alignment information between text and speech supervision, as well as phone-level duration prediction. MaskGCT is a two-stage model: in the first stage, the model uses text to predict semantic tokens extracted from a speech self-supervised learning (SSL) model, and in the second stage, the model predicts acoustic tokens conditioned on these semantic tokens. MaskGCT follows the mask-and-predict learning paradigm. During training, MaskGCT learns to predict masked semantic or acoustic tokens based on given conditions and prompts. During inference, the model generates tokens of a specified length in a parallel manner. Experiments with 100K hours of in-the-wild speech demonstrate that MaskGCT outperforms the current state-of-the-art zero-shot TTS systems in terms of quality, similarity, and intelligibility. Audio samples are available at https://maskgct.github.io/. We release our code and model checkpoints at https://github.com/open-mmlab/Amphion/blob/main/models/tts/maskgct.
URLs: https://maskgct.github.io/., https://github.com/open-mmlab/Amphion/blob/main/models/tts/maskgct.
Authors: Renjie Wu, Hu Wang, Hsiang-Ting Chen, Gustavo Carneiro
Abstract: During multimodal model training and testing, certain data modalities may be absent due to sensor limitations, cost constraints, privacy concerns, or data loss, negatively affecting performance. Multimodal learning techniques designed to handle missing modalities can mitigate this by ensuring model robustness even when some modalities are unavailable. This survey reviews recent progress in Multimodal Learning with Missing Modality (MLMM), focusing on deep learning methods. It provides the first comprehensive survey that covers the motivation and distinctions between MLMM and standard multimodal learning setups, followed by a detailed analysis of current methods, applications, and datasets, concluding with challenges and future directions.
Authors: Andreas Opedal, Eleanor Chodroff, Ryan Cotterell, Ethan Gotlieb Wilcox
Abstract: We present a new perspective on how readers integrate context during real-time language comprehension. Our proposals build on surprisal theory, which posits that the processing effort of a linguistic unit (e.g., a word) is an affine function of its in-context information content. We first observe that surprisal is only one out of many potential ways that a contextual predictor can be derived from a language model. Another one is the pointwise mutual information (PMI) between a unit and its context, which turns out to yield the same predictive power as surprisal when controlling for unigram frequency. Moreover, both PMI and surprisal are correlated with frequency. This means that neither PMI nor surprisal contains information about context alone. In response to this, we propose a technique where we project surprisal onto the orthogonal complement of frequency, yielding a new contextual predictor that is uncorrelated with frequency. Our experiments show that the proportion of variance in reading times explained by context is a lot smaller when context is represented by the orthogonalized predictor. From an interpretability standpoint, this indicates that previous studies may have overstated the role that context has in predicting reading times.
Authors: Ehsan Kabir, Md. Arafat Kabir, Austin R. J. Downey, Jason D. Bakos, David Andrews, Miaoqing Huang
Abstract: Transformer neural networks (TNNs) are being applied across a widening range of application domains, including natural language processing (NLP), machine translation, and computer vision (CV). Their popularity is largely attributed to the exceptional performance of their multi-head self-attention blocks when analyzing sequential data and extracting features. To date, there are limited hardware accelerators tailored for this mechanism, which is the first step before designing an accelerator for a complete model. This paper proposes \textit{FAMOUS}, a flexible hardware accelerator for dense multi-head attention (MHA) computation of TNNs on field-programmable gate arrays (FPGAs). It is optimized for high utilization of processing elements and on-chip memories to improve parallelism and reduce latency. An efficient tiling of large matrices has been employed to distribute memory and computing resources across different modules on various FPGA platforms. The design is evaluated on Xilinx Alveo U55C and U200 data center cards containing Ultrascale+ FPGAs. Experimental results are presented that show that it can attain a maximum throughput, number of parallel attention heads, embedding dimension and tile size of 328 (giga operations/second (GOPS)), 8, 768 and 64 respectively on the U55C. Furthermore, it is 3.28$\times$ and 2.6$\times$ faster than the Intel Xeon Gold 5220R CPU and NVIDIA V100 GPU respectively. It is also 1.3$\times$ faster than the fastest state-of-the-art FPGA-based accelerator.
Authors: Tiansheng Huang, Sihao Hu, Fatih Ilhan, Selim Furkan Tekin, Ling Liu
Abstract: Recent research demonstrates that the nascent fine-tuning-as-a-service business model exposes serious safety concerns -- fine-tuning over a few harmful data uploaded by the users can compromise the safety alignment of the model. The attack, known as harmful fine-tuning, has raised a broad research interest among the community. However, as the attack is still new, \textbf{we observe from our miserable submission experience that there are general misunderstandings within the research community.} We in this paper aim to clear some common concerns for the attack setting, and formally establish the research problem. Specifically, we first present the threat model of the problem, and introduce the harmful fine-tuning attack and its variants. Then we systematically survey the existing literature on attacks/defenses/mechanical analysis of the problem. Finally, we outline future research directions that might contribute to the development of the field. Additionally, we present a list of questions of interest, which might be useful to refer to when reviewers in the peer review process question the realism of the experiment/attack/defense setting. A curated list of relevant papers is maintained and made accessible at: \url{https://github.com/git-disl/awesome_LLM-harmful-fine-tuning-papers}.
URLs: https://github.com/git-disl/awesome_LLM-harmful-fine-tuning-papers
Authors: Brandon L. Annesi, Enrico M. Malatesta, Francesco Zamponi
Abstract: We analyze the problem of storing random pattern-label associations using two classes of continuous non-convex weights models, namely the perceptron with negative margin and an infinite-width two-layer neural network with non-overlapping receptive fields and generic activation function. Using a full-RSB ansatz we compute the exact value of the SAT/UNSAT transition. Furthermore, in the case of the negative perceptron we show that the overlap distribution of typical states displays an overlap gap (a disconnected support) in certain regions of the phase diagram defined by the value of the margin and the density of patterns to be stored. This implies that some recent theorems that ensure convergence of Approximate Message Passing (AMP) based algorithms to capacity are not applicable. Finally, we show that Gradient Descent is not able to reach the maximal capacity, irrespectively of the presence of an overlap gap for typical states. This finding, similarly to what occurs in binary weight models, suggests that gradient-based algorithms are biased towards highly atypical states, whose inaccessibility determines the algorithmic threshold.
Authors: Ludovica Schaerf
Abstract: The latent space of image generative models is a multi-dimensional space of compressed hidden visual knowledge. Its entity captivates computer scientists, digital artists, and media scholars alike. Latent space has become an aesthetic category in AI art, inspiring artistic techniques such as the latent space walk, exemplified by the works of Mario Klingemann and others. It is also viewed as cultural snapshots, encoding rich representations of our visual world. This paper proposes a double view of the latent space, as a multi-dimensional archive of culture and as a multi-dimensional space of potentiality. The paper discusses disentanglement as a method to elucidate the double nature of the space and as an interpretative direction to exploit its organization in human terms. The paper compares the role of disentanglement as potentiality to that of conditioning, as imagination, and confronts this interpretation with the philosophy of Deleuzian potentiality and Hume's imagination. Lastly, this paper notes the difference between traditional generative models and recent architectures.
Authors: Abhijit Manatkar, Ashlesha Akella, Parthivi Gupta, Krishnasuri Narayanam
Abstract: Discovering meaningful insights from a large dataset, known as Exploratory Data Analysis (EDA), is a challenging task that requires thorough exploration and analysis of the data. Automated Data Exploration (ADE) systems use goal-oriented methods with Large Language Models and Reinforcement Learning towards full automation. However, these methods require human involvement to anticipate goals that may limit insight extraction, while fully automated systems demand significant computational resources and retraining for new datasets. We introduce QUIS, a fully automated EDA system that operates in two stages: insight generation (ISGen) driven by question generation (QUGen). The QUGen module generates questions in iterations, refining them from previous iterations to enhance coverage without human intervention or manually curated examples. The ISGen module analyzes data to produce multiple relevant insights in response to each question, requiring no prior training and enabling QUIS to adapt to new datasets.
Authors: Pengrui Quan, Xiaomin Ouyang, Jeya Vikranth Jeyakumar, Ziqi Wang, Yang Xing, Mani Srivastava
Abstract: Effective processing, interpretation, and management of sensor data have emerged as a critical component of cyber-physical systems. Traditionally, processing sensor data requires profound theoretical knowledge and proficiency in signal-processing tools. However, recent works show that Large Language Models (LLMs) have promising capabilities in processing sensory data, suggesting their potential as copilots for developing sensing systems. To explore this potential, we construct a comprehensive benchmark, SensorBench, to establish a quantifiable objective. The benchmark incorporates diverse real-world sensor datasets for various tasks. The results show that while LLMs exhibit considerable proficiency in simpler tasks, they face inherent challenges in processing compositional tasks with parameter selections compared to engineering experts. Additionally, we investigate four prompting strategies for sensor processing and show that self-verification can outperform all other baselines in 48% of tasks. Our study provides a comprehensive benchmark and prompting analysis for future developments, paving the way toward an LLM-based sensor processing copilot.
Authors: Tsz Ting Chung, Leyang Cui, Lemao Liu, Xinting Huang, Shuming Shi, Dit-Yan Yeung
Abstract: Large Language Models (LLMs) have demonstrated impressive capabilities in a wide range of natural language processing tasks when leveraging in-context learning. To mitigate the additional computational and financial costs associated with in-context learning, several prompt compression methods have been proposed to compress the in-context learning prompts. Despite their success, these methods face challenges with transferability due to model-specific compression, or rely on external training data, such as GPT-4. In this paper, we investigate the ability of LLMs to develop a unified compression method that discretizes uninformative tokens, utilizing a self-supervised pre-training technique. By introducing a small number of parameters during the continual pre-training, the proposed Selection-p produces a probability for each input token, indicating whether to preserve or discard it. Experiments show Selection-p achieves state-of-the-art performance across numerous classification tasks, achieving compression rates of up to 10 times while experiencing only a marginal 0.8% decrease in performance. Moreover, it exhibits superior transferability to different models compared to prior work. Additionally, we further analyze how Selection-p helps maintain performance on in-context learning with long contexts.
Authors: Erik Arakelyan, Pasquale Minervini, Pat Verga, Patrick Lewis, Isabelle Augenstein
Abstract: Modern Question Answering (QA) and Reasoning approaches based on Large Language Models (LLMs) commonly use prompting techniques, such as Chain-of-Thought (CoT), assuming the resulting generation will have a more granular exploration and reasoning over the question space and scope. However, such methods struggle with generating outputs that are faithful to the intermediate chain of reasoning produced by the model. On the other end of the spectrum, neuro-symbolic methods such as Faithful CoT (F-CoT) propose to combine LLMs with external symbolic solvers. While such approaches boast a high degree of faithfulness, they usually require a model trained for code generation and struggle with tasks that are ambiguous or hard to formalise strictly. We introduce $\textbf{F}$aithful $\textbf{L}$ogic-$\textbf{A}$ided $\textbf{R}$easoning and $\textbf{E}$xploration ($\textbf{FLARE}$), a novel interpretable approach for traversing the problem space using task decompositions. We use the LLM to plan a solution, soft-formalise the query into facts and predicates using a logic programming code and simulate that code execution using an exhaustive multi-hop search over the defined space. Our method allows us to compute the faithfulness of the reasoning process w.r.t. the generated code and analyse the steps of the multi-hop search without relying on external solvers. Our methods achieve SOTA results on $\mathbf{7}$ out of $\mathbf{9}$ diverse reasoning benchmarks. We also show that model faithfulness positively correlates with overall performance and further demonstrate that $\textbf{FLARE}$ allows pinpointing the decisive factors sufficient for and leading to the correct answer with optimal reasoning during the multi-hop search.
Authors: Hongduan Tian, Feng Liu, Zhanke Zhou, Tongliang Liu, Chengqi Zhang, Bo Han
Abstract: In cross-domain few-shot classification (CFC), recent works mainly focus on adapting a simple transformation head on top of a frozen pre-trained backbone with few labeled data to project embeddings into a task-specific metric space where classification can be performed by measuring similarities between image instance and prototype representations. Technically, an assumption implicitly adopted in such a framework is that the prototype and image instance embeddings share the same representation transformation. However, in this paper, we find that there naturally exists a gap, which resembles the modality gap, between the prototype and image instance embeddings extracted from the frozen pre-trained backbone, and simply applying the same transformation during the adaptation phase constrains exploring the optimal representations and shrinks the gap between prototype and image representations. To solve this problem, we propose a simple yet effective method, contrastive prototype-image adaptation (CoPA), to adapt different transformations respectively for prototypes and images similarly to CLIP by treating prototypes as text prompts. Extensive experiments on Meta-Dataset demonstrate that CoPA achieves the state-of-the-art performance more efficiently. Meanwhile, further analyses also indicate that CoPA can learn better representation clusters, enlarge the gap, and achieve minimal validation loss at the enlarged gap.
Authors: Dwipam Katariya, Juan Manuel Origgi, Yage Wang, Thomas Caputo
Abstract: Users engage with financial services companies through multiple channels, often interacting with mobile applications, web platforms, call centers, and physical locations to service their accounts. The resulting interactions are recorded at heterogeneous temporal resolutions across these domains. This multi-channel data can be combined and encoded to create a comprehensive representation of the customer's journey for accurate intent prediction. This demands sequential learning solutions. NMT transformers achieve state-of-the-art sequential representation learning by encoding context and decoding for the next best action to represent long-range dependencies. However, three major challenges exist while combining multi-domain sequences within an encode-decoder transformers architecture for intent prediction applications: a) aligning sequences with different sampling rates b) learning temporal dynamics across multi-variate, multi-domain sequences c) combining dynamic and static sequences. We propose an encoder-decoder transformer model to address these challenges for contextual and sequential intent prediction in financial servicing applications. Our experiments show significant improvement over the existing tabular method.
Authors: Maxx Richard Rahman, Ruoxuan Liu, Wolfgang Maass
Abstract: Anomaly detection in clinical time-series holds significant potential in identifying suspicious patterns in different biological parameters. In this paper, we propose a targeted method that incorporates the clinical domain knowledge into LLMs to improve their ability to detect anomalies. We introduce the Metabolism Pathway-driven Prompting (MPP) method, which integrates the information about metabolic pathways to better capture the structural and temporal changes in biological samples. We applied our method for doping detection in sports, focusing on steroid metabolism, and evaluated using real-world data from athletes. The results show that our method improves anomaly detection performance by leveraging metabolic context, providing a more nuanced and accurate prediction of suspicious samples in athletes' profiles.
Authors: Zihao Zeng, Bokai Lin, Tianqi Hou, Hao Zhang, Zhijie Deng
Abstract: The KV-Cache technique has become the standard for the inference of large language models (LLMs). It caches states of self-attention to avoid recomputation. Yet, it is widely criticized that KV-Cache can become a bottleneck of the LLM inference system, especially when confronted with ultra-large models and long-context queries. A natural remedy is to discard the KV-Cache for less important tokens, with StreamingLLM as an example, but the used static eviction strategies cannot flexibly adapt to varying contexts. Remedies like H2O leverage accumulative attention scores to perform dynamic eviction but suffer from the attention bias issue in capturing contextual information. This paper bridges this gap by devising a parameterized KV-Cache eviction mechanism, dubbed as Attention-Gate, which accepts the whole context as input and yields eviction flags for each token to realize in-context eviction. The subsequent self-attention module proceeds according to the flags and only the KV states for the remaining tokens need to be cached. The Attention-Gates can vary among different heads and layers and be trivially plugged into pre-trained LLMs, tuned by cost-effective continual pre-training or supervised fine-tuning objectives to acquire what to discard. The computational and memory overhead introduced by Attention-Gates is minimal. Our method is validated across multiple tasks, demonstrating both efficiency and adaptability. After a highly efficient continual pre-training, it achieves higher average accuracy and evicts more tokens compared to traditional training-free methods. In supervised fine-tuning, it not only evicts many tokens but also outperforms LoRA-finetuned LLMs on some datasets, such as RTE, where it improves accuracy by 13.9% while evicting 62.8% of tokens, showing that effective eviction of redundant tokens can even enhance performance.
Authors: Lai Wei, Ambuj Tewari, Michael A. Cianfrocco
Abstract: We introduce a novel extension of the contextual bandit problem, where new sets of arms can be requested with stochastic time delays and associated costs. In this setting, the learner can select multiple arms from a decision set, with each selection taking one unit of time. The problem is framed as a special case of semi-Markov decision processes (SMDPs). The arm contexts, request times, and costs are assumed to follow an unknown distribution. We consider the regret of an online learning algorithm with respect to the optimal policy that achieves the maximum average reward. By leveraging the Bellman optimality equation, we design algorithms that can effectively select arms and determine the appropriate time to request new arms, thereby minimizing their regret. Under the realizability assumption, we analyze the proposed algorithms and demonstrate that their regret upper bounds align with established results in the contextual bandit literature. We validate the algorithms through experiments on simulated data and a movie recommendation dataset, showing that their performance is consistent with theoretical analyses.
Authors: Mihail Stoian, Alexander van Renen, Jan Kobiolka, Ping-Lin Kuo, Josif Grabocka, Andreas Kipf
Abstract: The growing adoption of data lakes for managing relational data necessitates efficient, open storage formats that provide high scan performance and competitive compression ratios. While existing formats achieve fast scans through lightweight encoding techniques, they have reached a plateau in terms of minimizing storage footprint. Recently, correlation-aware compression schemes have been shown to reduce file sizes further. Yet, current approaches either incur significant scan overheads or require manual specification of correlations, limiting their practicability. We present $\texttt{Virtual}$, a framework that integrates seamlessly with existing open formats to automatically leverage data correlations, achieving substantial compression gains while having minimal scan performance overhead. Experiments on data-gov datasets show that $\texttt{Virtual}$ reduces file sizes by up to 40% compared to Apache Parquet.
Authors: Gaoyang Pang, Wanchun Liu, Dusit Niyato, Daniel Quevedo, Branka Vucetic, Yonghui Li
Abstract: Wireless Human-Machine Collaboration (WHMC) represents a critical advancement for Industry 5.0, enabling seamless interaction between humans and machines across geographically distributed systems. As the WHMC systems become increasingly important for achieving complex collaborative control tasks, ensuring their stability is essential for practical deployment and long-term operation. Stability analysis certifies how the closed-loop system will behave under model randomness, which is essential for systems operating with wireless communications. However, the fundamental stability analysis of the WHMC systems remains an unexplored challenge due to the intricate interplay between the stochastic nature of wireless communications, dynamic human operations, and the inherent complexities of control system dynamics. This paper establishes a fundamental WHMC model incorporating dual wireless loops for machine and human control. Our framework accounts for practical factors such as short-packet transmissions, fading channels, and advanced HARQ schemes. We model human control lag as a Markov process, which is crucial for capturing the stochastic nature of human interactions. Building on this model, we propose a stochastic cycle-cost-based approach to derive a stability condition for the WHMC system, expressed in terms of wireless channel statistics, human dynamics, and control parameters. Our findings are validated through extensive numerical simulations and a proof-of-concept experiment, where we developed and tested a novel wireless collaborative cart-pole control system. The results confirm the effectiveness of our approach and provide a robust framework for future research on WHMC systems in more complex environments.
Authors: J\'erome Eertmans, Enrico Maria Vittuci, Vittorio Degli Esposti, Laurent Jacques, Claude Oestges
Abstract: With the increasing presence of dynamic scenarios, such as Vehicle-to-Vehicle communications, radio propagation modeling tools must adapt to the rapidly changing nature of the radio channel. Recently, both Differentiable and Dynamic Ray Tracing frameworks have emerged to address these challenges. However, there is often confusion about how these approaches differ and which one should be used in specific contexts. In this paper, we provide an overview of these two techniques and a comparative analysis against two state-of-the-art tools: 3DSCAT from UniBo and Sionna from NVIDIA. To provide a more precise characterization of the scope of these methods, we introduce a novel simulation-based metric, the Multipath Lifetime Map, which enables the evaluation of spatial and temporal coherence in radio channels only based on the geometrical description of the environment. Finally, our metrics are evaluated on a classic urban street canyon scenario, yielding similar results to those obtained from measurement campaigns.