Authors: Daniel Andr\'es D\'iaz-Pach\'on, H. Renata Gallegos, Ola H\"ossjer, J. Sunil Rao
Abstract: In this paper, we study learning and knowledge acquisition (LKA) of an agent about a proposition that is either true or false. We use a Bayesian approach, where the agent receives data to update his beliefs about the proposition according to a posterior distribution. The LKA is formulated in terms of active information, with data representing external or exogenous information that modifies the agent's beliefs. It is assumed that data provide details about a number of features that are relevant to the proposition. We show that this leads to a Gibbs distribution posterior, which is in maximum entropy relative to the prior, conditioned on the side constraints that the data provide in terms of the features. We demonstrate that full learning is sometimes not possible and full knowledge acquisition is never possible when the number of extracted features is too small. We also distinguish between primary learning (receiving data about features of relevance for the proposition) and secondary learning (receiving data about the learning of another agent). We argue that this type of secondary learning does not represent true knowledge acquisition. Our results have implications for statistical learning algorithms, and we claim that such algorithms do not always generate true knowledge. The theory is illustrated with several examples.
Authors: Sophia Nguyen, Beihao Zhou, Yi Ding, Sihang Liu
Abstract: In this work, we study LLMs from a carbon emission perspective, addressing both operational and embodied emissions, and paving the way for sustainable LLM serving. We characterize the performance and energy of LLaMA with 1B, 3B, and 7B parameters using two Nvidia GPU types, a latest-generation RTX6000 Ada and an older-generation T4. We analytically model operational carbon emissions based on energy consumption and carbon intensities from three grid regions -- each representing a different energy source mix, and embodied carbon emissions based on chip area and memory size. Our characterization and modeling provide us with an in-depth understanding of the performance, energy, and carbon emissions of LLM serving. Our findings highlight the potential for optimizing sustainable LLM serving systems by considering both operational and embodied carbon emissions simultaneously.
Authors: You Zhou, Changsheng You, Kaibin Huang
Abstract: Rare events, despite their infrequency, often carry critical information and require immediate attentions in mission-critical applications such as autonomous driving, healthcare, and industrial automation. The data-intensive nature of these tasks and their need for prompt responses, combined with designing edge AI (or edge inference), pose significant challenges in systems and techniques. Existing edge inference approaches often suffer from communication bottlenecks due to high-dimensional data transmission and fail to provide timely responses to rare events, limiting their effectiveness for mission-critical applications in the sixth-generation (6G) mobile networks. To overcome these challenges, we propose a channel-adaptive, event-triggered edge-inference framework that prioritizes efficient rare-event processing. Central to this framework is a dual-threshold, multi-exit architecture, which enables early local inference for rare events detected locally while offloading more complex rare events to edge servers for detailed classification. To further enhance the system's performance, we developed a channel-adaptive offloading policy paired with an online algorithm to dynamically determine the optimal confidence thresholds for controlling offloading decisions. The associated optimization problem is solved by reformulating the original non-convex function into an equivalent strongly convex one. Using deep neural network classifiers and real medical datasets, our experiments demonstrate that the proposed framework not only achieves superior rare-event classification accuracy, but also effectively reduces communication overhead, as opposed to existing edge-inference approaches.
Authors: Guhan Sivakumar
Abstract: This paper explores the application of Hidden Markov Models (HMM) and Long Short-Term Memory (LSTM) neural networks for economic forecasting, focusing on predicting CPI inflation rates. The study explores a new approach that integrates HMM-derived hidden states and means as additional features for LSTM modeling, aiming to enhance the interpretability and predictive performance of the models. The research begins with data collection and preprocessing, followed by the implementation of the HMM to identify hidden states representing distinct economic conditions. Subsequently, LSTM models are trained using the original and augmented data sets, allowing for comparative analysis and evaluation. The results demonstrate that incorporating HMM-derived data improves the predictive accuracy of LSTM models, particularly in capturing complex temporal patterns and mitigating the impact of volatile economic conditions. Additionally, the paper discusses the implementation of Integrated Gradients for model interpretability and provides insights into the economic dynamics reflected in the forecasting outcomes.
Authors: Jianfeng Xu, Congcong Liu, Xiaoying Tan, Xiaojie Zhu, Anpeng Wu, Huan Wan, Weijun Kong, Chun Li, Hu Xu, Kun Kuang, Fei Wu
Abstract: To address the growing size of AI model training data and the lack of a universal data selection methodology-factors that significantly drive up training costs -- this paper presents the General Information Metrics Evaluation (GIME) method. GIME leverages general information metrics from Objective Information Theory (OIT), including volume, delay, scope, granularity, variety, duration, sampling rate, aggregation, coverage, distortion, and mismatch to optimize dataset selection for training purposes. Comprehensive experiments conducted across diverse domains, such as CTR Prediction, Civil Case Prediction, and Weather Forecasting, demonstrate that GIME effectively preserves model performance while substantially reducing both training time and costs. Additionally, applying GIME within the Judicial AI Program led to a remarkable 39.56% reduction in total model training expenses, underscoring its potential to support efficient and sustainable AI development.
Authors: Xi Yu, Tiejun Lv, Weicai Li, Wei Ni, Dusit Niyato, Ekram Hossain
Abstract: Multi-task semantic communication can serve multiple learning tasks using a shared encoder model. Existing models have overlooked the intricate relationships between features extracted during an encoding process of tasks. This paper presents a new graph attention inter-block (GAI) module to the encoder/transmitter of a multi-task semantic communication system, which enriches the features for multiple tasks by embedding the intermediate outputs of encoding in the features, compared to the existing techniques. The key idea is that we interpret the outputs of the intermediate feature extraction blocks of the encoder as the nodes of a graph to capture the correlations of the intermediate features. Another important aspect is that we refine the node representation using a graph attention mechanism to extract the correlations and a multi-layer perceptron network to associate the node representations with different tasks. Consequently, the intermediate features are weighted and embedded into the features transmitted for executing multiple tasks at the receiver. Experiments demonstrate that the proposed model surpasses the most competitive and publicly available models by 11.4% on the CityScapes 2Task dataset and outperforms the established state-of-the-art by 3.97% on the NYU V2 3Task dataset, respectively, when the bandwidth ratio of the communication channel (i.e., compression level for transmission over the channel) is as constrained as 1 12 .
Authors: Yannis Y. He
Abstract: In the realm of neural architecture design, achieving high performance is largely reliant on the manual expertise of researchers. Despite the emergence of Neural Architecture Search (NAS) as a promising technique for automating this process, current NAS methods still require human input to expand the search space and cannot generate new architectures. This paper explores the potential of Transformers in comprehending neural architectures and their performance, with the objective of establishing the foundation for utilizing Transformers to generate novel networks. We propose the Token-based Architecture Transformer (TART), which predicts neural network performance without the need to train candidate networks. TART attains state-of-the-art performance on the DeepNets-1M dataset for performance prediction tasks without edge information, indicating the potential of Transformers to aid in discovering novel and high-performing neural architectures.
Authors: Antoine Ledent, Peng Liu
Abstract: Balancing predictive power and interpretability has long been a challenging research area, particularly in powerful yet complex models like neural networks, where nonlinearity obstructs direct interpretation. This paper introduces a novel approach to constructing an explainable neural network that harmonizes predictiveness and explainability. Our model, termed SparXnet, is designed as a linear combination of a sparse set of jointly learned features, each derived from a different trainable function applied to a single 1-dimensional input feature. Leveraging the ability to learn arbitrarily complex relationships, our neural network architecture enables automatic selection of a sparse set of important features, with the final prediction being a linear combination of rescaled versions of these features. We demonstrate the ability to select significant features while maintaining comparable predictive performance and direct interpretability through extensive experiments on synthetic and real-world datasets. We also provide theoretical analysis on the generalization bounds of our framework, which is favorably linear in the number of selected features and only logarithmic in the number of input features. We further lift any dependence of sample complexity on the number of parameters or the architectural details under very mild conditions. Our research paves the way for further research on sparse and explainable neural networks with guarantee.
Authors: Keng Hou Leong (Victor), Yuxuan Xiu (Victor), Wai Kin (Victor), Chan
Abstract: The representations of conditional entropy and conditional mutual information are significant in explaining the unique effects among variables. While previous studies based on conditional contrastive sampling have effectively removed information regarding discrete sensitive variables, they have not yet extended their scope to continuous cases. This paper introduces Information Subtraction, a framework designed to generate representations that preserve desired information while eliminating the undesired. We implement a generative-based architecture that outputs these representations by simultaneously maximizing an information term and minimizing another. With its flexibility in disentangling information, we can iteratively apply Information Subtraction to represent arbitrary information components between continuous variables, thereby explaining the various relationships that exist between them. Our results highlight the representations' ability to provide semantic features of conditional entropy. By subtracting sensitive and domain-specific information, our framework demonstrates effective performance in fair learning and domain generalization. The code for this paper is available at https://github.com/jh-liang/Information-Subtraction
Authors: Masahiro Matsumoto, Abu Saleh Musa Miah, Nobuyoshi Asai, Jungpil Shin
Abstract: Parkinson's disease (PD), the second most common neurodegenerative disorder, is characterized by dopaminergic neuron loss and the accumulation of abnormal synuclein. PD presents both motor and non-motor symptoms that progressively impair daily functioning. The severity of these symptoms is typically assessed using the MDS-UPDRS rating scale, which is subjective and dependent on the physician's experience. Additionally, PD shares symptoms with other neurodegenerative diseases, such as progressive supranuclear palsy (PSP) and multiple system atrophy (MSA), complicating accurate diagnosis. To address these diagnostic challenges, we propose a machine learning-based system for differential diagnosis of PD, PSP, MSA, and healthy controls (HC). This system utilizes a kinematic feature-based hierarchical feature extraction and selection approach. Initially, 18 kinematic features are extracted, including two newly proposed features: Thumb-to-index vector velocity and acceleration, which provide insights into motor control patterns. In addition, 41 statistical features were extracted here from each kinematic feature, including some new approaches such as Average Absolute Change, Rhythm, Amplitude, Frequency, Standard Deviation of Frequency, and Slope. Feature selection is performed using One-way ANOVA to rank features, followed by Sequential Forward Floating Selection (SFFS) to identify the most relevant ones, aiming to reduce the computational complexity. The final feature set is used for classification, achieving a classification accuracy of 66.67% for each dataset and 88.89% for each patient, with particularly high performance for the MSA and HC groups using the SVM algorithm. This system shows potential as a rapid and accurate diagnostic tool in clinical practice, though further data collection and refinement are needed to enhance its reliability.
Authors: Hwa Hui Tew, Gaoxuan Li, Fan Ding, Xuewen Luo, Junn Yong Loo, Chee-Ming Ting, Ze Yang Ding, Chee Pin Tan
Abstract: Soft sensing of hard-to-measure variables is often crucial in industrial processes. Current practices rely heavily on conventional modeling techniques that show success in improving accuracy. However, they overlook the non-linear nature, dynamics characteristics, and non-Euclidean dependencies between complex process variables. To tackle these challenges, we present a framework known as a Knowledge discovery graph Attention Network for effective Soft sensing (KANS). Unlike the existing deep learning soft sensor models, KANS can discover the intrinsic correlations and irregular relationships between the multivariate industrial processes without a predefined topology. First, an unsupervised graph structure learning method is introduced, incorporating the cosine similarity between different sensor embedding to capture the correlations between sensors. Next, we present a graph attention-based representation learning that can compute the multivariate data parallelly to enhance the model in learning complex sensor nodes and edges. To fully explore KANS, knowledge discovery analysis has also been conducted to demonstrate the interpretability of the model. Experimental results demonstrate that KANS significantly outperforms all the baselines and state-of-the-art methods in soft sensing performance. Furthermore, the analysis shows that KANS can find sensors closely related to different process variables without domain knowledge, significantly improving soft sensing accuracy.
Authors: Hwa Hui Tew, Fan Ding, Gaoxuan Li, Junn Yong Loo, Chee-Ming Ting, Ze Yang Ding, Chee Pin Tan
Abstract: Higher-order sensor networks are more accurate in characterizing the nonlinear dynamics of sensory time-series data in modern industrial settings by allowing multi-node connections beyond simple pairwise graph edges. In light of this, we propose a deep spatio-temporal hypergraph convolutional neural network for soft sensing (ST-HCSS). In particular, our proposed framework is able to construct and leverage a higher-order graph (hypergraph) to model the complex multi-interactions between sensor nodes in the absence of prior structural knowledge. To capture rich spatio-temporal relationships underlying sensor data, our proposed ST-HCSS incorporates stacked gated temporal and hypergraph convolution layers to effectively aggregate and update hypergraph information across time and nodes. Our results validate the superiority of ST-HCSS compared to existing state-of-the-art soft sensors, and demonstrates that the learned hypergraph feature representations aligns well with the sensor data correlations. The code is available at https://github.com/htew0001/ST-HCSS.git
Authors: Radha Nagarajan, Marco Scutari
Abstract: Modeling the associations between real world entities from their multivariate cross-sectional profiles can provide cues into the concerted working of these entities as a system. Several techniques have been proposed for deciphering these associations including constraint-based Bayesian structure learning (BSL) algorithms that model them as directed acyclic graphs. Benchmarking these algorithms have typically focused on assessing the variation in performance measures such as sensitivity as a function of the dimensionality represented by the number of nodes in the DAG, and sample size. The present study elucidates the importance of network topology in benchmarking exercises. More specifically, it investigates variations in sensitivity across distinct network topologies while constraining the nodes, edges, and sample-size to be identical, eliminating these as potential confounders. Sensitivity of three popular constraint-based BSL algorithms (Peter-Clarke, Grow-Shrink, Incremental Association Markov Blanket) in learning the network structure from multivariate cross-sectional profiles sampled from network models with sub-linear, linear, and super-linear DAG topologies generated using preferential attachment is investigated. Results across linear and nonlinear models revealed statistically significant $(\alpha=0.05)$ decrease in sensitivity estimates from sub-linear to super-linear topology constitutively across the three algorithms. These results are demonstrated on networks with nodes $(N_{nods}=48,64)$, noise strengths $(\sigma =3,6)$ and sample size $(N = 2^{10})$. The findings elucidate the importance of accommodating the network topology in constraint-based BSL benchmarking exercises.
Authors: Aditya Prakash
Abstract: Graph classification plays a pivotal role in various domains, including pathology, where images can be represented as graphs.In this domain, images can be represented as graphs, where nodes might represent individual nuclei, and edges capture the spatial or functional relationships between them. Often, the overall label of the graph, such as a cancer type or disease state, is determined by patterns within smaller, localized regions of the image. This work introduces a weakly-supervised graph classification framework leveraging two subgraph extraction techniques: (1) Sliding-window approach (2) BFS-based approach. Subgraphs are processed using a Graph Attention Network (GAT), which employs attention mechanisms to identify the most informative subgraphs for classification. Weak supervision is achieved by propagating graph-level labels to subgraphs, eliminating the need for detailed subgraph annotations.
Authors: Aashish Cheruvu, Nathaniel Rigoni
Abstract: This paper presents a novel deep learning-based approach named RealDiffFusionNet incorporating Neural Controlled Differential Equations (Neural CDE) - time series models that are robust in handling irregularly sampled data - and multi-head attention to align relevant multimodal context (image data, time invariant data, etc.) at each time point. Long short-term memory (LSTM) models were also used as a baseline. Two different datasets were used: a data from the Open-Source Imaging Consortium (OSIC) containing structured time series data of demographics and lung function with a baseline CT scan of the lungs and the second from the Alzheimer's Disease Neuroimaging Initiative (ADNI) containing a series of MRI scans along with demographics, physical examinations, and cognitive assessment data. An ablation study was performed to understand the role of CDEs, multimodal data, attention fusion, and interpolation strategies on model performance. When the baseline models were evaluated, the use of multimodal data resulted in an improvement in Neural CDE performance, with a lower test RMSE. Additionally, the performance of multimodal Neural CDE was also superior to multimodal LSTM. In the attention-based architectures, fusion through concatenation and rectilinear interpolation were found to improve model performance. The performance of the proposed RealDiffFusionNet was found to be superior (0.2570) to all models. For the ADNI dataset, between the Neural-CDE and LSTM models trained only on the structured data, the test RMSE were comparable (0.471 for LSTM vs. 0.4581 Neural-CDE). Furthermore, the addition of image features from patients' MRI series resulted in an improvement in performance, with a lower test RMSE (0.4372 with multimodal vs 0.4581 with structured data). RealDiffFusionNet has shown promise in utilizing CDEs and multimodal data to accurately predict disease progression.
Authors: Ziwei Zheng, Junyao Zhao, Le Yang, Lijun He, Fan Li
Abstract: With the integration of an additional modality, large vision-language models (LVLMs) exhibit greater vulnerability to safety risks (e.g., jailbreaking) compared to their language-only predecessors. Although recent studies have devoted considerable effort to the post-hoc alignment of LVLMs, the inner safety mechanisms remain largely unexplored. In this paper, we discover that internal activations of LVLMs during the first token generation can effectively identify malicious prompts across different attacks. This inherent safety perception is governed by sparse attention heads, which we term ``safety heads." Further analysis reveals that these heads act as specialized shields against malicious prompts; ablating them leads to higher attack success rates, while the model's utility remains unaffected. By locating these safety heads and concatenating their activations, we construct a straightforward but powerful malicious prompt detector that integrates seamlessly into the generation process with minimal extra inference overhead. Despite its simple structure of a logistic regression model, the detector surprisingly exhibits strong zero-shot generalization capabilities. Experiments across various prompt-based attacks confirm the effectiveness of leveraging safety heads to protect LVLMs. Code is available at \url{https://github.com/Ziwei-Zheng/SAHs}.
Authors: Tianyu Cheng, Qun Chen
Abstract: Deep clustering is an essential task in modern artificial intelligence, aiming to partition a set of data samples into a given number of homogeneous groups (i.e., clusters). Even though many Deep Neural Network (DNN) backbones and clustering strategies have been proposed for the task, achieving increasingly improved performance, deep clustering remains very challenging due to the lack of accurately labeled samples. In this paper, we propose a novel approach of deep clustering via community detection. It initializes clustering by detecting many communities, and then gradually expands clusters by community merging. Compared with the existing clustering strategies, community detection factors in the new perspective of cluster network analysis. As a result, it has the inherent benefit of high pseudo-label purity, which is critical to the performance of self-supervision. We have validated the efficacy of the proposed approach on benchmark image datasets. Our extensive experiments have shown that it can effectively improve the SOTA performance. Our ablation study also demonstrates that the new network perspective can effectively improve community pseudo-label purity, resulting in improved clustering performance.
Authors: David S\'anchez Pedroche, Daniel Amigo, Jes\'us Garc\'ia, Jose M. Molina
Abstract: This paper proposes a data preparation process for managing real-world kinematic data and detecting fishing vessels. The solution is a binary classification that classifies ship trajectories into either fishing or non-fishing ships. The data used are characterized by the typical problems found in classic data mining applications using real-world data, such as noise and inconsistencies. The two classes are also clearly unbalanced in the data, a problem which is addressed using algorithms that resample the instances. For classification, a series of features are extracted from spatiotemporal data that represent the trajectories of the ships, available from sequences of Automatic Identification System (AIS) reports. These features are proposed for the modelling of ship behavior but, because they do not contain context-related information, the classification can be applied in other scenarios. Experimentation shows that the proposed data preparation process is useful for the presented classification problem. In addition, positive results are obtained using minimal information.
Authors: Christopher Burger, Charles Walter, Thai Le, Lingwei Chen
Abstract: Recent work has investigated the concept of adversarial attacks on explainable AI (XAI) in the NLP domain with a focus on examining the vulnerability of local surrogate methods such as Lime to adversarial perturbations or small changes on the input of a machine learning (ML) model. In such attacks, the generated explanation is manipulated while the meaning and structure of the original input remain similar under the ML model. Such attacks are especially alarming when XAI is used as a basis for decision making (e.g., prescribing drugs based on AI medical predictors) or for legal action (e.g., legal dispute involving AI software). Although weaknesses across many XAI methods have been shown to exist, the reasons behind why remain little explored. Central to this XAI manipulation is the similarity measure used to calculate how one explanation differs from another. A poor choice of similarity measure can lead to erroneous conclusions about the stability or adversarial robustness of an XAI method. Therefore, this work investigates a variety of similarity measures designed for text-based ranked lists referenced in related work to determine their comparative suitability for use. We find that many measures are overly sensitive, resulting in erroneous estimates of stability. We then propose a weighting scheme for text-based data that incorporates the synonymity between the features within an explanation, providing more accurate estimates of the actual weakness of XAI methods to adversarial examples.
Authors: Evan R. Antoniuk, Peggy Li, Nathan Keilbart, Stephen Weitzner, Bhavya Kailkhura, Anna M. Hiszpanski
Abstract: Although generative models hold promise for discovering molecules with optimized desired properties, they often fail to suggest synthesizable molecules that improve upon the known molecules seen in training. We find that a key limitation is not in the molecule generation process itself, but in the poor generalization capabilities of molecular property predictors. We tackle this challenge by creating an active-learning, closed-loop molecule generation pipeline, whereby molecular generative models are iteratively refined on feedback from quantum chemical simulations to improve generalization to new chemical space. Compared against other generative model approaches, only our active learning approach generates molecules with properties that extrapolate beyond the training data (reaching up to 0.44 standard deviations beyond the training data range) and out-of-distribution molecule classification accuracy is improved by 79%. By conditioning molecular generation on thermodynamic stability data from the active-learning loop, the proportion of stable molecules generated is 3.5x higher than the next-best model.
Authors: Abhishek Srinivasan, Varun Singapuri Ravi, Juan Carlos Andresen, Anders Holst
Abstract: The complexity of modern electro-mechanical systems require the development of sophisticated diagnostic methods like anomaly detection capable of detecting deviations. Conventional anomaly detection approaches like signal processing and statistical modelling often struggle to effectively handle the intricacies of complex systems, particularly when dealing with multi-variate signals. In contrast, neural network-based anomaly detection methods, especially Auto-Encoders, have emerged as a compelling alternative, demonstrating remarkable performance. However, Auto-Encoders exhibit inherent opaqueness in their decision-making processes, hindering their practical implementation at scale. Addressing this opacity is essential for enhancing the interpretability and trustworthiness of anomaly detection models. In this work, we address this challenge by employing a feature selector to select features and counterfactual explanations to give a context to the model output. We tested this approach on the SKAB benchmark dataset and an industrial time-series dataset. The gradient based counterfactual explanation approach was evaluated via validity, sparsity and distance measures. Our experimental findings illustrate that our proposed counterfactual approach can offer meaningful and valuable insights into the model decision-making process, by explaining fewer signals compared to conventional approaches. These insights enhance the trustworthiness and interpretability of anomaly detection models.
Authors: Mehrdad Moghimi, Hyejin Ku
Abstract: In domains such as finance, healthcare, and robotics, managing worst-case scenarios is critical, as failure to do so can lead to catastrophic outcomes. Distributional Reinforcement Learning (DRL) provides a natural framework to incorporate risk sensitivity into decision-making processes. However, existing approaches face two key limitations: (1) the use of fixed risk measures at each decision step often results in overly conservative policies, and (2) the interpretation and theoretical properties of the learned policies remain unclear. While optimizing a static risk measure addresses these issues, its use in the DRL framework has been limited to the simple static CVaR risk measure. In this paper, we present a novel DRL algorithm with convergence guarantees that optimizes for a broader class of static Spectral Risk Measures (SRM). Additionally, we provide a clear interpretation of the learned policy by leveraging the distribution of returns in DRL and the decomposition of static coherent risk measures. Extensive experiments demonstrate that our model learns policies aligned with the SRM objective, and outperforms existing risk-neutral and risk-sensitive DRL models in various settings.
Authors: Ming Yin, Mengdi Wang, Yu-Xiang Wang
Abstract: This article reviews the recent advances on the statistical foundation of reinforcement learning (RL) in the offline and low-adaptive settings. We will start by arguing why offline RL is the appropriate model for almost any real-life ML problems, even if they have nothing to do with the recent AI breakthroughs that use RL. Then we will zoom into two fundamental problems of offline RL: offline policy evaluation (OPE) and offline policy learning (OPL). It may be surprising to people that tight bounds for these problems were not known even for tabular and linear cases until recently. We delineate the differences between worst-case minimax bounds and instance-dependent bounds. We also cover key algorithmic ideas and proof techniques behind near-optimal instance-dependent methods in OPE and OPL. Finally, we discuss the limitations of offline RL and review a burgeoning problem of \emph{low-adaptive exploration} which addresses these limitations by providing a sweet middle ground between offline and online RL.
Authors: Jin Li, Kleanthis Malialis, Stelios G. Vrachimis, Marios M. Polycarpou
Abstract: Water Distribution Networks (WDNs) are vital infrastructures, and contamination poses serious public health risks. Harmful substances can interact with disinfectants like chlorine, making chlorine monitoring essential for detecting contaminants. However, chlorine sensors often become unreliable and require frequent calibration. This study introduces the Dual-Threshold Anomaly and Drift Detection (AD&DD) method, an unsupervised approach combining a dual-threshold drift detection mechanism with an LSTM-based Variational Autoencoder(LSTM-VAE) for real-time contamination detection. Tested on two realistic WDNs, AD&DD effectively identifies anomalies with sensor offsets as concept drift, and outperforms other methods. A proposed decentralized architecture enables accurate contamination detection and localization by deploying AD&DD on selected nodes.
Authors: Ishaan Maitra, Raymond Lin, Eric Chen, Jon Donnelly, Sanja \v{S}\'cepanovi\'c, Cynthia Rudin
Abstract: Health outcomes depend on complex environmental and sociodemographic factors whose effects change over location and time. Only recently has fine-grained spatial and temporal data become available to study these effects, namely the MEDSAT dataset of English health, environmental, and sociodemographic information. Leveraging this new resource, we use a variety of variable importance techniques to robustly identify the most informative predictors across multiple health outcomes. We then develop an interpretable machine learning framework based on Generalized Additive Models (GAMs) and Multiscale Geographically Weighted Regression (MGWR) to analyze both local and global spatial dependencies of each variable on various health outcomes. Our findings identify NO2 as a global predictor for asthma, hypertension, and anxiety, alongside other outcome-specific predictors related to occupation, marriage, and vegetation. Regional analyses reveal local variations with air pollution and solar radiation, with notable shifts during COVID. This comprehensive approach provides actionable insights for addressing health disparities, and advocates for the integration of interpretable machine learning in public health.
Authors: Chien-Ping Lu
Abstract: As large-scale AI models expand, training becomes costlier and sustaining progress grows harder. Classical scaling laws (e.g., Kaplan et al. (2020), Hoffmann et al. (2022)) predict training loss from a static compute budget yet neglect time and efficiency, prompting the question: how can we balance ballooning GPU fleets with rapidly improving hardware and algorithms? We introduce the relative-loss equation, a time- and efficiency-aware framework that extends classical AI scaling laws. Our model shows that, without ongoing efficiency gains, advanced performance could demand millennia of training or unrealistically large GPU fleets. However, near-exponential progress remains achievable if the "efficiency-doubling rate" parallels Moore's Law. By formalizing this race to efficiency, we offer a quantitative roadmap for balancing front-loaded GPU investments with incremental improvements across the AI stack. Empirical trends suggest that sustained efficiency gains can push AI scaling well into the coming decade, providing a new perspective on the diminishing returns inherent in classical scaling.
Authors: Ying Chen, Jiajing Chen, Yijie Weng, ChiaHua Chang, Dezhi Yu, Guanbiao Lin
Abstract: Membership inference attacks have emerged as a significant privacy concern in the training of deep learning models, where attackers can infer whether a data point was part of the training set based on the model's outputs. To address this challenge, we propose a novel defense mechanism, AdaMixup. AdaMixup employs adaptive mixup techniques to enhance the model's robustness against membership inference attacks by dynamically adjusting the mixup strategy during training. This method not only improves the model's privacy protection but also maintains high performance. Experimental results across multiple datasets demonstrate that AdaMixup significantly reduces the risk of membership inference attacks while achieving a favorable trade-off between defensive efficiency and model accuracy. This research provides an effective solution for data privacy protection and lays the groundwork for future advancements in mixup training methods.
Authors: Jianwei Wang, Kai Wang, Ying Zhang, Wenjie Zhang, Xiwei Xu, Xuemin Lin
Abstract: Missing data imputation, which aims to impute the missing values in the raw datasets to achieve the completeness of datasets, is crucial for modern data-driven models like large language models (LLMs) and has attracted increasing interest over the past decades. Despite its importance, existing solutions for missing data imputation either 1) only support numerical and categorical data or 2) show an unsatisfactory performance due to their design prioritizing text data and the lack of key properties for tabular data imputation. In this paper, we propose UnIMP, a Unified IMPutation framework that leverages LLM and high-order message passing to enhance the imputation of mixed-type data including numerical, categorical, and text data. Specifically, we first introduce a cell-oriented hypergraph to model the table. We then propose BiHMP, an efficient Bidirectional High-order Message-Passing network to aggregate global-local information and high-order relationships on the constructed hypergraph while capturing the inter-column heterogeneity and intra-column homogeneity. To effectively and efficiently align the capacity of the LLM with the information aggregated by BiHMP, we introduce Xfusion, which, together with BiHMP, acts as adapters for the LLM. We follow a pre-training and fine-tuning pipeline to train UnIMP, integrating two optimizations: chunking technique, which divides tables into smaller chunks to enhance efficiency; and progressive masking technique, which gradually adapts the model to learn more complex data patterns. Both theoretical proofs and empirical experiments on 10 real world datasets highlight the superiority of UnIMP over existing techniques.
Authors: Zhongyi Zhou, Yaxin Peng, Pin Yi, Minjie Zhu, Chaomin Shen
Abstract: Continual Learning enables models to learn and adapt to new tasks while retaining prior knowledge.Introducing new tasks, however, can naturally lead to feature entanglement across tasks, limiting the model's capability to distinguish between new domain data.In this work, we propose a method called Feature Realignment through Experts on hyperSpHere in Continual Learning (Fresh-CL). By leveraging predefined and fixed simplex equiangular tight frame (ETF) classifiers on a hypersphere, our model improves feature separation both intra and inter tasks.However, the projection to a simplex ETF shifts with new tasks, disrupting structured feature representation of previous tasks and degrading performance. Therefore, we propose a dynamic extension of ETF through mixture of experts, enabling adaptive projections onto diverse subspaces to enhance feature representation.Experiments on 11 datasets demonstrate a 2\% improvement in accuracy compared to the strongest baseline, particularly in fine-grained datasets, confirming the efficacy of combining ETF and MoE to improve feature distinction in continual learning scenarios.
Authors: Hua Zheng, Wei Xie, Ilya O. Ryzhov, Keilung Choy
Abstract: This paper presents a novel methodological framework, called the Actor-Simulator, that incorporates the calibration of digital twins into model-based reinforcement learning for more effective control of stochastic systems with complex nonlinear dynamics. Traditional model-based control often relies on restrictive structural assumptions (such as linear state transitions) and fails to account for parameter uncertainty in the model. These issues become particularly critical in industries such as biopharmaceutical manufacturing, where process dynamics are complex and not fully known, and only a limited amount of data is available. Our approach jointly calibrates the digital twin and searches for an optimal control policy, thus accounting for and reducing model error. We balance exploration and exploitation by using policy performance as a guide for data collection. This dual-component approach provably converges to the optimal policy, and outperforms existing methods in extensive numerical experiments based on the biopharmaceutical manufacturing domain.
Authors: Zhongwei Wang, Tong Wu, Zhiyong Chen, Liang Qian, Yin Xu, Meixia Tao
Abstract: Federated semi-supervised learning (FSSL) is primarily challenged by two factors: the scarcity of labeled data across clients and the non-independent and identically distribution (non-IID) nature of data among clients. In this paper, we propose a novel approach, diffusion model-based data synthesis aided FSSL (DDSA-FSSL), which utilizes a diffusion model (DM) to generate synthetic data, bridging the gap between heterogeneous local data distributions and the global data distribution. In DDSA-FSSL, clients address the challenge of the scarcity of labeled data by employing a federated learning-trained classifier to perform pseudo labeling for unlabeled data. The DM is then collaboratively trained using both labeled and precision-optimized pseudo-labeled data, enabling clients to generate synthetic samples for classes that are absent in their labeled datasets. This process allows clients to generate more comprehensive synthetic datasets aligned with the global distribution. Extensive experiments conducted on multiple datasets and varying non-IID distributions demonstrate the effectiveness of DDSA-FSSL, e.g., it improves accuracy from 38.46% to 52.14% on CIFAR-10 datasets with 10% labeled data.
Authors: Yangze Zhou, Guoxin Lin, Gonghao Zhang, Yi Wang
Abstract: Meteorological factors (MF) are crucial in day-ahead load forecasting as they significantly influence the electricity consumption behaviors of consumers. Numerous studies have incorporated MF into the load forecasting model to achieve higher accuracy. Selecting MF from one representative location or the averaged MF as the inputs of the forecasting model is a common practice. However, the difference in MF collected in various locations within a region may be significant, which poses a challenge in selecting the appropriate MF from numerous locations. A representation learning framework is proposed to extract geo-distributed MF while considering their spatial relationships. In addition, this paper employs the Shapley value in the graph-based model to reveal connections between MF collected in different locations and loads. To reduce the computational complexity of calculating the Shapley value, an acceleration method is adopted based on Monte Carlo sampling and weighted linear regression. Experiments on two real-world datasets demonstrate that the proposed method improves the day-ahead forecasting accuracy, especially in extreme scenarios such as the "accumulation temperature effect" in summer and "sudden temperature change" in winter. We also find a significant correlation between the importance of MF in different locations and the corresponding area's GDP and mainstay industry.
Authors: Zongwei Li, Lianghao Xia, Hua Hua, Shijie Zhang, Shuangyang Wang, Chao Huang
Abstract: Recent advances in Graph Neural Networks (GNNs) have revolutionized graph-structured data modeling, yet traditional GNNs struggle with complex heterogeneous structures prevalent in real-world scenarios. Despite progress in handling heterogeneous interactions, two fundamental challenges persist: noisy data significantly compromising embedding quality and learning performance, and existing methods' inability to capture intricate semantic transitions among heterogeneous relations, which impacts downstream predictions. To address these fundamental issues, we present the Heterogeneous Graph Diffusion Model (DiffGraph), a pioneering framework that introduces an innovative cross-view denoising strategy. This advanced approach transforms auxiliary heterogeneous data into target semantic spaces, enabling precise distillation of task-relevant information. At its core, DiffGraph features a sophisticated latent heterogeneous graph diffusion mechanism, implementing a novel forward and backward diffusion process for superior noise management. This methodology achieves simultaneous heterogeneous graph denoising and cross-type transition, while significantly simplifying graph generation through its latent-space diffusion capabilities. Through rigorous experimental validation on both public and industrial datasets, we demonstrate that DiffGraph consistently surpasses existing methods in link prediction and node classification tasks, establishing new benchmarks for robustness and efficiency in heterogeneous graph processing. The model implementation is publicly available at: https://github.com/HKUDS/DiffGraph.
Authors: Seyed Mahdi B. Azad, Zahra Padar, Gabriel Kalweit, Joschka Boedecker
Abstract: In this paper, we propose a novel method for learning reward functions directly from offline demonstrations. Unlike traditional inverse reinforcement learning (IRL), our approach decouples the reward function from the learner's policy, eliminating the adversarial interaction typically required between the two. This results in a more stable and efficient training process. Our reward function, called \textit{SR-Reward}, leverages successor representation (SR) to encode a state based on expected future states' visitation under the demonstration policy and transition dynamics. By utilizing the Bellman equation, SR-Reward can be learned concurrently with most reinforcement learning (RL) algorithms without altering the existing training pipeline. We also introduce a negative sampling strategy to mitigate overestimation errors by reducing rewards for out-of-distribution data, thereby enhancing robustness. This strategy inherently introduces a conservative bias into RL algorithms that employ the learned reward. We evaluate our method on the D4RL benchmark, achieving competitive results compared to offline RL algorithms with access to true rewards and imitation learning (IL) techniques like behavioral cloning. Moreover, our ablation studies on data size and quality reveal the advantages and limitations of SR-Reward as a proxy for true rewards.
Authors: Yahya Sowti Khiabani, Farris Atif, Chieh Hsu, Sven Stahlmann, Tobias Michels, Sebastian Kramer, Benedikt Heidrich, M. Saquib Sarfraz, Julian Merten, Faezeh Tafazzoli
Abstract: We propose a holistic approach for deploying Small Language Models (SLMs) as function-calling agents within vehicles as edge devices, offering a more flexible and robust alternative to traditional rule-based systems. By leveraging SLMs, we simplify vehicle control mechanisms and enhance the user experience. Given the in-vehicle hardware constraints, we apply state-of-the-art model compression techniques, including structured pruning, healing, and quantization, ensuring that the model fits within the resource limitations while maintaining acceptable performance. Our work focuses on optimizing a representative SLM, Microsoft's Phi-3 mini, and outlines best practices for enabling embedded models, including compression, task-specific fine-tuning, and vehicle integration. We demonstrate that, despite significant reduction in model size which removes up to 2 billion parameters from the original model, our approach preserves the model's ability to handle complex in-vehicle tasks accurately and efficiently. Furthermore, by executing the model in a lightweight runtime environment, we achieve a generation speed of 11 tokens per second, making real-time, on-device inference feasible without hardware acceleration. Our results demonstrate the potential of SLMs to transform vehicle control systems, enabling more intuitive interactions between users and their vehicles for an enhanced driving experience.
Authors: Yikai Zhang, Jiahe Lin, Fengpei Li, Songzhu Zheng, Anant Raj, Anderson Schneider, Yuriy Nevmyvaka
Abstract: In this work, we study the weighted empirical risk minimization (weighted ERM) schema, in which an additional data-dependent weight function is incorporated when the empirical risk function is being minimized. We show that under a general ``balanceable" Bernstein condition, one can design a weighted ERM estimator to achieve superior performance in certain sub-regions over the one obtained from standard ERM, and the superiority manifests itself through a data-dependent constant term in the error bound. These sub-regions correspond to large-margin ones in classification settings and low-variance ones in heteroscedastic regression settings, respectively. Our findings are supported by evidence from synthetic data experiments.
Authors: P. Barcel\'o, R. Cominetti, M. Morgado
Abstract: Feature attribution methods have become essential for explaining machine learning models. Many popular approaches, such as SHAP and Banzhaf values, are grounded in power indices from cooperative game theory, which measure the contribution of features to model predictions. This work studies the computational complexity of power indices beyond SHAP, addressing the conditions under which they can be computed efficiently. We identify a simple condition on power indices that ensures that computation is polynomially equivalent to evaluating expected values, extending known results for SHAP. We also introduce Bernoulli power indices, showing that their computation can be simplified to a constant number of expected value evaluations. Furthermore, we explore interaction power indices that quantify the importance of feature subsets, proving that their computation complexity mirrors that of individual features.
Authors: Ambroise Odonnat, Wassim Bouaziz, Vivien Cabannes
Abstract: Gradient descent is the method of choice for training large artificial intelligence systems. As these systems become larger, a better understanding of the mechanisms behind gradient training would allow us to alleviate compute costs and help steer these systems away from harmful behaviors. To that end, we suggest utilizing the circuit perspective brought forward by mechanistic interpretability. After laying out our intuition, we illustrate how it enables us to design a curriculum for efficient learning in a controlled setting. The code is available at \url{https://github.com/facebookresearch/pal}.
Authors: Alec S. Xu, Can Yaras, Peng Wang, Qing Qu
Abstract: Deep neural networks have attained remarkable success across diverse classification tasks. Recent empirical studies have shown that deep networks learn features that are linearly separable across classes. However, these findings often lack rigorous justifications, even under relatively simple settings. In this work, we address this gap by examining the linear separation capabilities of shallow nonlinear networks. Specifically, inspired by the low intrinsic dimensionality of image data, we model inputs as a union of low-dimensional subspaces (UoS) and demonstrate that a single nonlinear layer can transform such data into linearly separable sets. Theoretically, we show that this transformation occurs with high probability when using random weights and quadratic activations. Notably, we prove this can be achieved when the network width scales polynomially with the intrinsic dimension of the data rather than the ambient dimension. Experimental results corroborate these theoretical findings and demonstrate that similar linear separation properties hold in practical scenarios beyond our analytical scope. This work bridges the gap between empirical observations and theoretical understanding of the separation capacity of nonlinear networks, offering deeper insights into model interpretability and generalization.
Authors: Tamon Nakano Sebastian Baur Christoph R\"ath
Abstract: As an alternative approach for predicting complex dynamical systems where physics-based models are no longer reliable, reservoir computing (RC) has gained popularity. The hybrid approach is considered an interesting option for improving the prediction performance of RC. The idea is to combine a knowledge-based model (KBM) to support the fully data-driven RC prediction. There are three types of hybridization for RC, namely full hybrid (FH), input hybrid (IH) and output hybrid (OH), where it was shown that the latter one is superior in terms of the accuracy and the robustness for the prediction of low-dimensional chaotic systems. Here, we extend the formalism to the prediction of spatiotemporal patterns in two dimensions. To overcome the curse of dimensionality for this very high-dimensional case we employ the local states ansatz, where only a few locally adjacent time series are utilized for the RC-based prediction. Using simulation data from the Barkley model describing chaotic electrical wave propagation in cardiac tissue, we outline the formalism of high-dimensional hybrid RC and assess the performance of the different hybridization schemes. We find that all three methods (FH, IH and OH) perform better than reservoir only, where improvements are small when the model is very inaccurate. For small model errors and small reservoirs FH and OH perform nearly equally well and better than IH. Given the smaller CPU needs for OH and especially the better interpretability of it, OH is to be favored. For large reservoirs the performance of OH drops below that of FH and IH. Generally, it maybe advisable to test the three setups for a given application and select the best suited one that optimizes between the counteracting factors of prediction performance and CPU needs.
Authors: Chia-Yi Hsu, Yu-Lin Tsai, Yu Zhe, Yan-Lun Chen, Chih-Hsun Lin, Chia-Mu Yu, Yang Zhang, Chun-Ying Huang, Jun Sakuma
Abstract: Task arithmetic in large-scale pre-trained models enables flexible adaptation to diverse downstream tasks without extensive re-training. By leveraging task vectors (TVs), users can perform modular updates to pre-trained models through simple arithmetic operations like addition and subtraction. However, this flexibility introduces new security vulnerabilities. In this paper, we identify and evaluate the susceptibility of TVs to backdoor attacks, demonstrating how malicious actors can exploit TVs to compromise model integrity. By developing composite backdoors and eliminating redudant clean tasks, we introduce BadTV, a novel backdoor attack specifically designed to remain effective under task learning, forgetting, and analogies operations. Our extensive experiments reveal that BadTV achieves near-perfect attack success rates across various scenarios, significantly impacting the security of models using task arithmetic. We also explore existing defenses, showing that current methods fail to detect or mitigate BadTV. Our findings highlight the need for robust defense mechanisms to secure TVs in real-world applications, especially as TV services become more popular in machine-learning ecosystems.
Authors: Fatih Dinc, Ege Cirakman, Yiqi Jiang, Mert Yuksekgonul, Mark J. Schnitzer, Hidenori Tanaka
Abstract: \emph{Abrupt learning} is commonly observed in neural networks, where long plateaus in network performance are followed by rapid convergence to a desirable solution. Yet, despite its common occurrence, the complex interplay of task, network architecture, and learning rule has made it difficult to understand the underlying mechanisms. Here, we introduce a minimal dynamical system trained on a delayed-activation task and demonstrate analytically how even a one-dimensional system can exhibit abrupt learning through ghost points rather than bifurcations. Through our toy model, we show that the emergence of a ghost point destabilizes learning dynamics. We identify a critical learning rate that prevents learning through two distinct loss landscape features: a no-learning zone and an oscillatory minimum. Testing these predictions in recurrent neural networks (RNNs), we confirm that ghost points precede abrupt learning and accompany the destabilization of learning. We demonstrate two complementary remedies: lowering the model output confidence prevents the network from getting stuck in no-learning zones, while increasing trainable ranks beyond task requirements (\textit{i.e.}, adding sloppy parameters) provides more stable learning trajectories. Our model reveals a bifurcation-free mechanism for abrupt learning and illustrates the importance of both deliberate uncertainty and redundancy in stabilizing learning dynamics.
Authors: Robert Joseph George, David Pitt, Jiawei Zhao, Jean Kossaifi, Cheng Luo, Yuandong Tian, Anima Anandkumar
Abstract: We present Tensor-GaLore, a novel method for efficient training of neural networks with higher-order tensor weights. Many models, particularly those used in scientific computing, employ tensor-parameterized layers to capture complex, multidimensional relationships. When scaling these methods to high-resolution problems makes memory usage grow intractably, and matrix based optimization methods lead to suboptimal performance and compression. We propose to work directly in the high-order space of the complex tensor parameter space using a tensor factorization of the gradients during optimization. We showcase its effectiveness on Fourier Neural Operators (FNOs), a class of models crucial for solving partial differential equations (PDE) and prove the theory of it. Across various PDE tasks like the Navier Stokes and Darcy Flow equations, Tensor-GaLore achieves substantial memory savings, reducing optimizer memory usage by up to 75%. These substantial memory savings across AI for science demonstrate Tensor-GaLore's potential.
Authors: Markus J. Buehler
Abstract: We present an approach to modifying Transformer architectures by integrating graph-aware relational reasoning into the attention mechanism, merging concepts from graph neural networks and language modeling. Building on the inherent connection between attention and graph theory, we reformulate the Transformer's attention mechanism as a graph operation and propose Graph-Aware Isomorphic Attention. This method leverages advanced graph modeling strategies, including Graph Isomorphism Networks (GIN) and Principal Neighborhood Aggregation (PNA), to enrich the representation of relational structures. Our approach captures complex dependencies and generalizes across tasks, as evidenced by a reduced generalization gap and improved learning performance. Additionally, we expand the concept of graph-aware attention to introduce Sparse GIN-Attention, a fine-tuning approach that employs sparse GINs. By interpreting attention matrices as sparse adjacency graphs, this technique enhances the adaptability of pre-trained foundational models with minimal computational overhead, endowing them with graph-aware capabilities. Sparse GIN-Attention fine-tuning achieves improved training dynamics and better generalization compared to alternative methods like low-rank adaption (LoRA). We discuss latent graph-like structures within traditional attention mechanisms, offering a new lens through which Transformers can be understood. By evolving Transformers as hierarchical GIN models for relational reasoning. This perspective suggests profound implications for foundational model development, enabling the design of architectures that dynamically adapt to both local and global dependencies. Applications in bioinformatics, materials science, language modeling, and beyond could benefit from this synthesis of relational and sequential data modeling, setting the stage for interpretable and generalizable modeling strategies.
Authors: Zaikang Lin, Sei Chang, Aaron Zweig, Elham Azizi, David A. Knowles
Abstract: Modern high-throughput biological datasets with thousands of perturbations provide the opportunity for large-scale discovery of causal graphs that represent the regulatory interactions between genes. Numerous methods have been proposed to infer a directed acyclic graph (DAG) corresponding to the underlying gene regulatory network (GRN) that captures causal gene relationships. However, existing models have restrictive assumptions (e.g. linearity, acyclicity), limited scalability, and/or fail to address the dynamic nature of biological processes such as cellular differentiation. We propose PerturbODE, a novel framework that incorporates biologically informative neural ordinary differential equations (neural ODEs) to model cell state trajectories under perturbations and derive the causal GRN from the neural ODE's parameters. We demonstrate PerturbODE's efficacy in trajectory prediction and GRN inference across simulated and real over-expression datasets.
Authors: Xingwu Sun, Shuaipeng Li, Ruobing Xie, Weidong Han, Kan Wu, Zhen Yang, Yixing Li, An Wang, Shuai Li, Jinbao Xue, Yu Cheng, Yangyu Tao, Zhanhui Kang, Chengzhong Xu, Di Wang, Jie Jiang
Abstract: Low-precision training is considered an effective strategy for reducing both training and downstream inference costs. Previous scaling laws for precision mainly focus on integer quantization, which pay less attention to the constituents in floating-point quantization and thus cannot well fit the LLM losses in this scenario. In contrast, while floating-point quantization training is more commonly implemented in production, the research on it has been relatively superficial. In this paper, we thoroughly explore the effects of floating-point quantization targets, exponent bits, mantissa bits, and the calculation granularity of the scaling factor in floating-point quantization training performance of LLM models. While presenting an accurate floating-point quantization unified scaling law, we also provide valuable suggestions for the community: (1) Exponent bits contribute slightly more to the model performance than mantissa bits. We provide the optimal exponent-mantissa bit ratio for different bit numbers, which is available for future reference by hardware manufacturers; (2) We discover the formation of the critical data size in low-precision LLM training. Too much training data exceeding the critical data size will inversely bring in degradation of LLM performance; (3) The optimal floating-point quantization precision is directly proportional to the computational power, but within a wide computational power range, we estimate that the best cost-performance precision lies between 4-8 bits.
Authors: Yuchen Lin, Yong Zhang, Sihan Feng, Hong Zhao
Abstract: Advancing artificial intelligence demands a deeper understanding of the mechanisms underlying deep learning. Here, we propose a straightforward analysis framework based on the dynamics of learning models. Neurons are categorized into two modes based on whether their transformation functions preserve order. This categorization reveals how deep neural networks (DNNs) maximize information extraction by rationally allocating the proportion of neurons in different modes across deep layers. We further introduce the attraction basins of the training samples in both the sample vector space and the weight vector space to characterize the generalization ability of DNNs. This framework allows us to identify optimal depth and width configurations, providing a unified explanation for fundamental DNN behaviors such as the "flat minima effect," "grokking," and double descent phenomena. Our analysis extends to networks with depths up to 100 layers.
Authors: Zhiwei Yao, Yang Xu, Hongli Xu, Yunming Liao, Zuan Xie
Abstract: Deploying Large Language Models (LLMs) on resource-constrained (or weak) devices presents significant challenges due to limited resources and heterogeneous data distribution. To address the data concern, it is necessary to fine-tune LLMs using on-device private data for various downstream tasks. While Federated Learning (FL) offers a promising privacy-preserving solution, existing fine-tuning methods retain the original LLM size, leaving issues of high inference latency and excessive memory demands unresolved. Hence, we design FedSpine, an FL framework that combines Parameter- Efficient Fine-Tuning (PEFT) with structured pruning for efficient deployment of LLMs on resource-constrained devices. Specifically, FedSpine introduces an iterative process to prune and tune the parameters of LLMs. To mitigate the impact of device heterogeneity, an online Multi-Armed Bandit (MAB) algorithm is employed to adaptively determine different pruning ratios and LoRA ranks for heterogeneous devices without any prior knowledge of their computing and communication capabilities. As a result, FedSpine maintains higher inference accuracy while improving fine-tuning efficiency. Experimental results conducted on a physical platform with 80 devices demonstrate that FedSpine can speed up fine-tuning by 1.4$\times$-6.9$\times$ and improve final accuracy by 0.4%-4.5% under the same sparsity level compared to other baselines.
Authors: Ramin Zarei-Sabzevar, Ahad Harati
Abstract: This paper proposes a novel Deep Positive-Negative Prototype (DPNP) model that combines prototype-based learning (PbL) with discriminative methods to improve class compactness and separability in deep neural networks. While PbL traditionally emphasizes interpretability by classifying samples based on their similarity to representative prototypes, it struggles with creating optimal decision boundaries in complex scenarios. Conversely, discriminative methods effectively separate classes but often lack intuitive interpretability. Toward exploiting advantages of these two approaches, the suggested DPNP model bridges between them by unifying class prototypes with weight vectors, thereby establishing a structured latent space that enables accurate classification using interpretable prototypes alongside a properly learned feature representation. Based on this central idea of unified prototype-weight representation, Deep Positive Prototype (DPP) is formed in the latent space as a representative for each class using off-the-shelf deep networks as feature extractors. Then, rival neighboring class DPPs are treated as implicit negative prototypes with repulsive force in DPNP, which push away DPPs from each other. This helps to enhance inter-class separation without the need for any extra parameters. Hence, through a novel loss function that integrates cross-entropy, prototype alignment, and separation terms, DPNP achieves well-organized feature space geometry, maximizing intra-class compactness and inter-class margins. We show that DPNP can organize prototypes in nearly regular positions within feature space, such that it is possible to achieve competitive classification accuracy even in much lower-dimensional feature spaces. Experimental results on several datasets demonstrate that DPNP outperforms state-of-the-art models, while using smaller networks.
Authors: Zhengpeng Xie, Jiahang Cao, Qiang Zhang, Jianxiong Zhang, Changwei Wang, Renjing Xu
Abstract: Humans rely on high-level meta-representations to engage in abstract reasoning. In complex cognitive tasks, these meta-representations help individuals abstract general rules from experience. However, constructing such meta-representations from high-dimensional observations remains a longstanding challenge for reinforcement learning agents. For instance, a well-trained agent often fails to generalize to even minor variations of the same task, such as changes in background color, while humans can easily handle. In this paper, we build a bridge between meta-representation and generalization, showing that generalization performance benefits from meta-representation learning. We also hypothesize that deep mutual learning (DML) among agents can help them converge to meta-representations. Empirical results provide support for our theory and hypothesis. Overall, this work provides a new perspective on the generalization of deep reinforcement learning.
Authors: Assaf Lahiany, Yehudit Aperstein
Abstract: For many practical applications, a high computational cost of inference over deep network architectures might be unacceptable. A small degradation in the overall inference accuracy might be a reasonable price to pay for a significant reduction in the required computational resources. In this work, we describe a method for introducing "shortcuts" into the DNN feedforward inference process by skipping costly feedforward computations whenever possible. The proposed method is based on the previously described BranchyNet (Teerapittayanon et al., 2016) and the EEnet (Demir, 2019) architectures that jointly train the main network and early exit branches. We extend those methods by attaching branches to pre-trained models and, thus, eliminating the need to alter the original weights of the network. We also suggest a new branch architecture based on convolutional building blocks to allow enough training capacity when applied on large DNNs. The proposed architecture includes confidence heads that are used for predicting the confidence level in the corresponding early exits. By defining adjusted thresholds on these confidence extensions, we can control in real-time the amount of data exiting from each branch and the overall tradeoff between speed and accuracy of our model. In our experiments, we evaluate our method using image datasets (SVHN and CIFAR10) and several DNN architectures (ResNet, DenseNet, VGG) with varied depth. Our results demonstrate that the proposed method enables us to reduce the average inference computational cost and further controlling the tradeoff between the model accuracy and the computation cost.
Authors: Alexander Kozachinskiy, Tomasz Steifer
Abstract: We construct a 3-layer constant-dimension transformer, recognizing the parity language, where neither parameter matrices nor the positional encoding depend on the input length. This improves upon a construction of Chiang and Cholak who use a positional encoding, depending on the input length (but their construction has 2 layers).
Authors: Zherui Huang, Yicheng Liu, Chumeng Liang, Guanjie Zheng
Abstract: Traffic signal control (TSC) is an important and widely studied direction. Recently, reinforcement learning (RL) methods have been used to solve TSC problems and achieve superior performance over conventional TSC methods. However, applying RL methods to the real world is challenging due to the huge cost of experiments in real-world traffic environments. One possible solution is TSC domain adaptation, which adapts trained models to target environments and reduces the number of interactions and the training cost. However, existing TSC domain adaptation methods still face two major issues: the lack of consideration for differences across cities and the low utilization of multi-city data. To solve aforementioned issues, we propose an approach named Adaptive Modularized Model (AMM). By modularizing TSC problems and network models, we overcome the challenge of possible changes in environmental observations. We also aggregate multi-city experience through meta-learning. We conduct extensive experiments on different cities and show that AMM can achieve excellent performance with limited interactions in target environments and outperform existing methods. We also demonstrate the feasibility and generalizability of our method.
Authors: Lin Wang, Qing Li
Abstract: Graph condensation reduces the size of large graphs while preserving performance, addressing the scalability challenges of Graph Neural Networks caused by computational inefficiencies on large datasets. Existing methods often rely on bi-level optimization, requiring extensive GNN training and limiting their scalability. To address these issues, this paper proposes Graph Condensation via Gaussian Process (GCGP), a novel and computationally efficient approach to graph condensation. GCGP utilizes a Gaussian Process (GP), with the condensed graph serving as observations, to estimate the posterior distribution of predictions. This approach eliminates the need for the iterative and resource-intensive training typically required by GNNs. To enhance the capability of the GCGP in capturing dependencies between function values, we derive a specialized covariance function that incorporates structural information. This covariance function broadens the receptive field of input nodes by local neighborhood aggregation, thereby facilitating the representation of intricate dependencies within the nodes. To address the challenge of optimizing binary structural information in condensed graphs, Concrete random variables are utilized to approximate the binary adjacency matrix in a continuous counterpart. This relaxation process allows the adjacency matrix to be represented in a differentiable form, enabling the application of gradient-based optimization techniques to discrete graph structures. Experimental results show that the proposed GCGP method efficiently condenses large-scale graph data while preserving predictive performance, addressing the scalability and efficiency challenges. The implementation of our method is publicly available at https://github.com/WANGLin0126/GCGP.
Authors: Jiaping Wang, Simiao Zhang, Qiao-Chu He, Yifan Chen
Abstract: The machine learning and data science community has made significant while dispersive progress in accelerating transformer-based large language models (LLMs), and one promising approach is to replace the original causal attention in a generative pre-trained transformer (GPT) with \emph{exponentially decaying causal linear attention}. In this paper, we present LeetDecoding, which is the first Python package that provides a large set of computation routines for this fundamental operator. The launch of LeetDecoding was motivated by the current lack of (1) clear understanding of the complexity regarding this operator, (2) a comprehensive collection of existing computation methods (usually spread in seemingly unrelated fields), and (3) CUDA implementations for fast inference on GPU. LeetDecoding's design is easy to integrate with existing linear-attention LLMs, and allows for researchers to benchmark and evaluate new computation methods for exponentially decaying causal linear attention. The usage of LeetDecoding does not require any knowledge of GPU programming and the underlying complexity analysis, intentionally making LeetDecoding accessible to LLM practitioners. The source code of LeetDecoding is provided at \href{https://github.com/Computational-Machine-Intelligence/LeetDecoding}{this GitHub repository}, and users can simply install LeetDecoding by the command \texttt{pip install leet-decoding}.
URLs: https://github.com/Computational-Machine-Intelligence/LeetDecoding
Authors: Priyanshu Singh, Kapil Ahuja
Abstract: Clustering algorithms are fundamental tools in data analysis, with hierarchical methods being particularly valuable for their flexibility. Chameleon is a widely used hierarchical clustering algorithm that excels at identifying high-quality clusters of arbitrary shapes, sizes, and densities. Chameleon2 is the most recent variant that has demonstrated significant improvements, but suffers from critical failings and there are certain improvements that can be made. The first failure we address is that the complexity of Chameleon2 is claimed to be $O(n^2)$, while we demonstrate that it is actually $O(n^2\log{n})$, with $n$ being the number of data points. Furthermore, we suggest improvements to Chameleon2 that ensure that the complexity remains $O(n^2)$ with minimal to no loss of performance. The second failing of Chameleon2 is that it lacks transparency and it does not provide the fine-tuned algorithm parameters used to obtain the claimed results. We meticulously provide all such parameter values to enhance replicability. The improvement which we make in Chameleon2 is that we replace the exact $k$-NN search with an approximate $k$-NN search. This further reduces the algorithmic complexity down to $O(n\log{n})$ without any performance loss. Here, we primarily configure three approximate nearest neighbor search algorithms (Annoy, FLANN and NMSLIB) to align with the overarching Chameleon2 clustering framework. Experimental evaluations on standard benchmark datasets demonstrate that the proposed Chameleon2++ algorithm is more efficient, robust, and computationally optimal.
Authors: Amol Khanna, Chenyi Ling, Derek Everett, Edward Raff, Nathan Inkawhich
Abstract: Existing methods for out-of-distribution (OOD) detection use various techniques to produce a score, separate from classification, that determines how ``OOD'' an input is. Our insight is that OOD detection can be simplified by using a neural network architecture which can effectively merge classification and OOD detection into a single step. Radial basis function networks (RBFNs) inherently link classification confidence and OOD detection; however, these networks have lost popularity due to the difficult of training them in a multi-layer fashion. In this work, we develop a multi-layer radial basis function network (MLRBFN) which can be easily trained. To ensure that these networks are also effective for OOD detection, we develop a novel depression mechanism. We apply MLRBFNs as standalone classifiers and as heads on top of pretrained feature extractors, and find that they are competitive with commonly used methods for OOD detection. Our MLRBFN architecture demonstrates a promising new direction for OOD detection methods.
Authors: Saleh Ashkboos, Mahdi Nikdan, Soroush Tabesh, Roberto L. Castro, Torsten Hoefler, Dan Alistarh
Abstract: Quantized training of Large Language Models (LLMs) remains an open challenge, as maintaining accuracy while performing all matrix multiplications in low precision has proven difficult. This is particularly the case when fine-tuning pre-trained models, which often already have large weight and activation outlier values that render quantized optimization difficult. We present HALO, a novel quantization-aware training approach for Transformers that enables accurate and efficient low-precision training by combining 1) strategic placement of Hadamard rotations in both forward and backward passes, to mitigate outliers during the low-precision computation, 2) FSDP integration for low-precision communication, and 3) high-performance kernel support. Our approach ensures that all large matrix multiplications during the forward and backward passes are executed in lower precision. Applied to LLAMA-family models, HALO achieves near-full-precision-equivalent results during fine-tuning on various tasks, while delivering up to 1.31x end-to-end speedup for full fine-tuning on RTX 4090 GPUs. Our method supports both standard and parameter-efficient fine-tuning (PEFT) methods, both backed by efficient kernel implementations. Our results demonstrate the first practical approach to fully quantized LLM fine-tuning that maintains accuracy in FP8 precision, while delivering performance benefits.
Authors: David Restrepo, Chenwei Wu, Yueran Jia, Jaden K. Sun, Jack Gallifant, Catherine G. Bielick, Yugang Jia, Leo A. Celi
Abstract: Accurate imputation of missing laboratory values in electronic health records (EHRs) is critical to enable robust clinical predictions and reduce biases in AI systems in healthcare. Existing methods, such as variational autoencoders (VAEs) and decision tree-based approaches such as XGBoost, struggle to model the complex temporal and contextual dependencies in EHR data, mainly in underrepresented groups. In this work, we propose Lab-MAE, a novel transformer-based masked autoencoder framework that leverages self-supervised learning for the imputation of continuous sequential lab values. Lab-MAE introduces a structured encoding scheme that jointly models laboratory test values and their corresponding timestamps, enabling explicit capturing temporal dependencies. Empirical evaluation on the MIMIC-IV dataset demonstrates that Lab-MAE significantly outperforms the state-of-the-art baselines such as XGBoost across multiple metrics, including root mean square error (RMSE), R-squared (R2), and Wasserstein distance (WD). Notably, Lab-MAE achieves equitable performance across demographic groups of patients, advancing fairness in clinical predictions. We further investigate the role of follow-up laboratory values as potential shortcut features, revealing Lab-MAE's robustness in scenarios where such data is unavailable. The findings suggest that our transformer-based architecture, adapted to the characteristics of the EHR data, offers a foundation model for more accurate and fair clinical imputation models. In addition, we measure and compare the carbon footprint of Lab-MAE with the baseline XGBoost model, highlighting its environmental requirements.
Authors: Shivaram Kalyanakrishnan, Sheel Shah, Santhosh Kumar Guguloth
Abstract: Reinforcement learning (RL) enables an agent interacting with an unknown MDP $M$ to optimise its behaviour by observing transitions sampled from $M$. A natural entity that emerges in the agent's reasoning is $\widehat{M}$, the maximum likelihood estimate of $M$ based on the observed transitions. The well-known \textit{certainty-equivalence} method (CEM) dictates that the agent update its behaviour to $\widehat{\pi}$, which is an optimal policy for $\widehat{M}$. Not only is CEM intuitive, it has been shown to enjoy minimax-optimal sample complexity in some regions of the parameter space for PAC RL with a generative model~\citep{Agarwal2020GenModel}. A seemingly unrelated algorithm is the ``trajectory tree method'' (TTM)~\citep{Kearns+MN:1999}, originally developed for efficient decision-time planning in large POMDPs. This paper presents a theoretical investigation that stems from the surprising finding that CEM may indeed be viewed as an application of TTM. The qualitative benefits of this view are (1) new and simple proofs of sample complexity upper bounds for CEM, in fact under a (2) weaker assumption on the rewards than is prevalent in the current literature. Our analysis applies to both non-stationary and stationary MDPs. Quantitatively, we obtain (3) improvements in the sample-complexity upper bounds for CEM both for non-stationary and stationary MDPs, in the regime that the ``mistake probability'' $\delta$ is small. Additionally, we show (4) a lower bound on the sample complexity for finite-horizon MDPs, which establishes the minimax-optimality of our upper bound for non-stationary MDPs in the small-$\delta$ regime.
Authors: Simin Javaherian, Bryce Turney, Li Chen, Nian-Feng Tzeng
Abstract: Federated Learning (FL) has gained prominence as a decentralized machine learning paradigm, allowing clients to collaboratively train a global model while preserving data privacy. Despite its potential, FL faces significant challenges in heterogeneous environments, where varying client resources and capabilities can undermine overall system performance. Existing approaches primarily focus on maximizing global model accuracy, often at the expense of unfairness among clients and suboptimal system efficiency, particularly in non-IID (non-Independent and Identically Distributed) settings. In this paper, we introduce FLamma, a novel Federated Learning framework based on adaptive gamma-based Stackelberg game, designed to address the aforementioned limitations and promote fairness. Our approach allows the server to act as the leader, dynamically adjusting a decay factor while clients, acting as followers, optimally select their number of local epochs to maximize their utility. Over time, the server incrementally balances client influence, initially rewarding higher-contributing clients and gradually leveling their impact, driving the system toward a Stackelberg Equilibrium. Extensive simulations on both IID and non-IID datasets show that our method significantly improves fairness in accuracy distribution without compromising overall model performance or convergence speed, outperforming traditional FL baselines.
Authors: Arya Hatamian, Lionel Levine, Haniyeh Ehsani Oskouie, Majid Sarrafzadeh
Abstract: Having a sufficient quantity of quality data is a critical enabler of training effective machine learning models. Being able to effectively determine the adequacy of a dataset prior to training and evaluating a model's performance would be an essential tool for anyone engaged in experimental design or data collection. However, despite the need for it, the ability to prospectively assess data sufficiency remains an elusive capability. We report here on two experiments undertaken in an attempt to better ascertain whether or not basic descriptive statistical measures can be indicative of how effective a dataset will be at training a resulting model. Leveraging the effect size of our features, this work first explores whether or not a correlation exists between effect size, and resulting model performance (theorizing that the magnitude of the distinction between classes could correlate to a classifier's resulting success). We then explore whether or not the magnitude of the effect size will impact the rate of convergence of our learning rate, (theorizing again that a greater effect size may indicate that the model will converge more rapidly, and with a smaller sample size needed). Our results appear to indicate that this is not an effective heuristic for determining adequate sample size or projecting model performance, and therefore that additional work is still needed to better prospectively assess adequacy of data.
Authors: Anh Tu Ngo, Chuan Song Heng, Nandish Chattopadhyay, Anupam Chattopadhyay
Abstract: Deep Neural Networks (DNNs) have gained considerable traction in recent years due to the unparalleled results they gathered. However, the cost behind training such sophisticated models is resource intensive, resulting in many to consider DNNs to be intellectual property (IP) to model owners. In this era of cloud computing, high-performance DNNs are often deployed all over the internet so that people can access them publicly. As such, DNN watermarking schemes, especially backdoor-based watermarks, have been actively developed in recent years to preserve proprietary rights. Nonetheless, there lies much uncertainty on the robustness of existing backdoor watermark schemes, towards both adversarial attacks and unintended means such as fine-tuning neural network models. One reason for this is that no complete guarantee of robustness can be assured in the context of backdoor-based watermark. In this paper, we extensively evaluate the persistence of recent backdoor-based watermarks within neural networks in the scenario of fine-tuning, we propose/develop a novel data-driven idea to restore watermark after fine-tuning without exposing the trigger set. Our empirical results show that by solely introducing training data after fine-tuning, the watermark can be restored if model parameters do not shift dramatically during fine-tuning. Depending on the types of trigger samples used, trigger accuracy can be reinstated to up to 100%. Our study further explores how the restoration process works using loss landscape visualization, as well as the idea of introducing training data in fine-tuning stage to alleviate watermark vanishing.
Authors: Sirong Wu, Xi Luo, Junjie Liu, Yuhui Deng
Abstract: Although large models have shown a strong capacity to solve large-scale problems in many areas including natural language and computer vision, their voluminous parameters are hard to deploy in a real-time system due to computational and energy constraints. Addressing this, knowledge distillation through Teacher-Student architecture offers a sustainable pathway to compress the knowledge of large models into more manageable sizes without significantly compromising performance. To enhance the robustness and interpretability of this framework, it is critical to understand how individual training data impact model performance, which is an area that remains underexplored. We propose the \textbf{Knowledge Distillation with Adaptive Influence Weight (KD-AIF)} framework which leverages influence functions from robust statistics to assign weights to training data, grounded in the four key SAFE principles: Sustainability, Accuracy, Fairness, and Explainability. This novel approach not only optimizes distillation but also increases transparency by revealing the significance of different data. The exploration of various update mechanisms within the KD-AIF framework further elucidates its potential to significantly improve learning efficiency and generalization in student models, marking a step toward more explainable and deployable Large Models. KD-AIF is effective in knowledge distillation while also showing exceptional performance in semi-supervised learning with outperforms existing baselines and methods in multiple benchmarks (CIFAR-100, CIFAR-10-4k, SVHN-1k, and GLUE).
Authors: Vivek Myers, Catherine Ji, Benjamin Eysenbach
Abstract: We study goal-conditioned RL through the lens of generalization, but not in the traditional sense of random augmentations and domain randomization. Rather, we aim to learn goal-directed policies that generalize with respect to the horizon: after training to reach nearby goals (which are easy to learn), these policies should succeed in reaching distant goals (which are quite challenging to learn). In the same way that invariance is closely linked with generalization is other areas of machine learning (e.g., normalization layers make a network invariant to scale, and therefore generalize to inputs of varying scales), we show that this notion of horizon generalization is closely linked with invariance to planning: a policy navigating towards a goal will select the same actions as if it were navigating to a waypoint en route to that goal. Thus, such a policy trained to reach nearby goals should succeed at reaching arbitrarily-distant goals. Our theoretical analysis proves that both horizon generalization and planning invariance are possible, under some assumptions. We present new experimental results and recall findings from prior work in support of our theoretical results. Taken together, our results open the door to studying how techniques for invariance and generalization developed in other areas of machine learning might be adapted to achieve this alluring property.
Authors: Naichang Ke, Ryogo Tanaka, Yoshinobu Kawahara
Abstract: We consider an operator-based latent Markov representation of a stochastic nonlinear dynamical system, where the stochastic evolution of the latent state embedded in a reproducing kernel Hilbert space is described with the corresponding transfer operator, and develop a spectral method to learn this representation based on the theory of stochastic realization. The embedding may be learned simultaneously using reproducing kernels, for example, constructed with feed-forward neural networks. We also address the generalization of sequential state-estimation (Kalman filtering) in stochastic nonlinear systems, and of operator-based eigen-mode decomposition of dynamics, for the representation. Several examples with synthetic and real-world data are shown to illustrate the empirical characteristics of our methods, and to investigate the performance of our model in sequential state-estimation and mode decomposition.
Authors: Bowen Fan, Yuming Ai, Xunkai Li, Zhilin Guo, Rong-Hua Li, Guoren Wang
Abstract: Graph Machine Learning is essential for understanding and analyzing relational data. However, privacy-sensitive applications demand the ability to efficiently remove sensitive information from trained graph neural networks (GNNs), avoiding the unnecessary time and space overhead caused by retraining models from scratch. To address this issue, Graph Unlearning (GU) has emerged as a critical solution, with the potential to support dynamic graph updates in data management systems and enable scalable unlearning in distributed data systems while ensuring privacy compliance. Unlike machine unlearning in computer vision or other fields, GU faces unique difficulties due to the non-Euclidean nature of graph data and the recursive message-passing mechanism of GNNs. Additionally, the diversity of downstream tasks and the complexity of unlearning requests further amplify these challenges. Despite the proliferation of diverse GU strategies, the absence of a benchmark providing fair comparisons for GU, and the limited flexibility in combining downstream tasks and unlearning requests, have yielded inconsistencies in evaluations, hindering the development of this domain. To fill this gap, we present OpenGU, the first GU benchmark, where 16 SOTA GU algorithms and 37 multi-domain datasets are integrated, enabling various downstream tasks with 13 GNN backbones when responding to flexible unlearning requests. Based on this unified benchmark framework, we are able to provide a comprehensive and fair evaluation for GU. Through extensive experimentation, we have drawn $8$ crucial conclusions about existing GU methods, while also gaining valuable insights into their limitations, shedding light on potential avenues for future research.
Authors: Huiqiang Chen, Tianqing Zhu, Wanlei Zhou, Wei Zhao
Abstract: Federated Learning (FL) has gained significant attention as it facilitates collaborative machine learning among multiple clients without centralizing their data on a server. FL ensures the privacy of participating clients by locally storing their data, which creates new challenges in fairness. Traditional debiasing methods assume centralized access to sensitive information, rendering them impractical for the FL setting. Additionally, FL is more susceptible to fairness issues than centralized machine learning due to the diverse client data sources that may be associated with group information. Therefore, training a fair model in FL without access to client local data is important and challenging. This paper presents AFed, a straightforward yet effective framework for promoting group fairness in FL. The core idea is to circumvent restricted data access by learning the global data distribution. This paper proposes two approaches: AFed-G, which uses a conditional generator trained on the server side, and AFed-GAN, which improves upon AFed-G by training a conditional GAN on the client side. We augment the client data with the generated samples to help remove bias. Our theoretical analysis justifies the proposed methods, and empirical results on multiple real-world datasets demonstrate a substantial improvement in AFed over several baselines.
Authors: Xiwen Chen, Peijie Qiu, Wenhui Zhu, Huayu Li, Hao Wang, Aristeidis Sotiras, Yalin Wang, Abolfazl Razi
Abstract: Since its introduction, the transformer has shifted the development trajectory away from traditional models (e.g., RNN, MLP) in time series forecasting, which is attributed to its ability to capture global dependencies within temporal tokens. Follow-up studies have largely involved altering the tokenization and self-attention modules to better adapt Transformers for addressing special challenges like non-stationarity, channel-wise dependency, and variable correlation in time series. However, we found that the expressive capability of sequence representation is a key factor influencing Transformer performance in time forecasting after investigating several representative methods, where there is an almost linear relationship between sequence representation entropy and mean square error, with more diverse representations performing better. In this paper, we propose a novel attention mechanism with Sequence Complementors and prove feasible from an information theory perspective, where these learnable sequences are able to provide complementary information beyond current input to feed attention. We further enhance the Sequence Complementors via a diversification loss that is theoretically covered. The empirical evaluation of both long-term and short-term forecasting has confirmed its superiority over the recent state-of-the-art methods.
Authors: Ting Wang, Zhixin Zhou, Rui Luo
Abstract: Graph Neural Networks (GNNs) has been widely used in a variety of fields because of their great potential in representing graph-structured data. However, lacking of rigorous uncertainty estimations limits their application in high-stakes. Conformal Prediction (CP) can produce statistically guaranteed uncertainty estimates by using the classifier's probability estimates to obtain prediction sets, which contains the true class with a user-specified probability. In this paper, we propose a Rank-based CP during training framework to GNNs (RCP-GNN) for reliable uncertainty estimates to enhance the trustworthiness of GNNs in the node classification scenario. By exploiting rank information of the classifier's outcome, prediction sets with desired coverage rate can be efficiently constructed. The strategy of CP during training with differentiable rank-based conformity loss function is further explored to adapt prediction sets according to network topology information. In this way, the composition of prediction sets can be guided by the goal of jointly reducing inefficiency and probability estimation errors. Extensive experiments on several real-world datasets show that our model achieves any pre-defined target marginal coverage while significantly reducing the inefficiency compared with state-of-the-art methods.
Authors: Zijian Wang, Bin Wang, Mingwen Shao, Hongbo Dou, Boxiang Tao
Abstract: Hybrid action models are widely considered an effective approach to reinforcement learning (RL) modeling. The current mainstream method is to train agents under Parameterized Action Markov Decision Processes (PAMDPs), which performs well in specific environments. Unfortunately, these models either exhibit drastic low learning efficiency in complex PAMDPs or lose crucial information in the conversion between raw space and latent space. To enhance the learning efficiency and asymptotic performance of the agent, we propose a model-based RL (MBRL) algorithm, FLEXplore. FLEXplore learns a parameterized-action-conditioned dynamics model and employs a modified Model Predictive Path Integral control. Unlike conventional MBRL algorithms, we carefully design the dynamics loss function and reward smoothing process to learn a loose yet flexible model. Additionally, we use the variational lower bound to maximize the mutual information between the state and the hybrid action, enhancing the exploration effectiveness of the agent. We theoretically demonstrate that FLEXplore can reduce the regret of the rollout trajectory through the Wasserstein Metric under given Lipschitz conditions. Our empirical results on several standard benchmarks show that FLEXplore has outstanding learning efficiency and asymptotic performance compared to other baselines.
Authors: Xin Cao, Qinghua Tao, Yingjie Zhou, Lu Zhang, Le Zhang, Dongjin Song, Dapeng Oliver Wu, Ce Zhu
Abstract: Residential load forecasting (RLF) is crucial for resource scheduling in power systems. Most existing methods utilize all given load records (dense data) to indiscriminately extract the dependencies between historical and future time series. However, there exist important regular patterns residing in the event-related associations among different appliances (sparse knowledge), which have yet been ignored.In this paper, we propose an Event-Response Knowledge Guided approach (ERKG) for RLF by incorporating the estimation of electricity usage events for different appliances, mining event-related sparse knowledge from the load series. With ERKG, the event-response estimation enables portraying the electricity consumption behaviors of residents, revealing regular variations in appliance operational states.To be specific, ERKG consists of knowledge extraction and guidance: i) a forecasting model is designed for the electricity usage events by estimating appliance operational states, aiming to extract the event-related sparse knowledge; ii) a novel knowledge-guided mechanism is established by fusing such state estimates of the appliance events into the RLF model, which can give particular focuses on the patterns of users' electricity consumption behaviors.Notably, ERKG can flexibly serve as a plug-in module to boost the capability of existing forecasting models by leveraging event response. In numerical experiments, extensive comparisons and ablation studies have verified the effectiveness of our ERKG, e.g., over 8% MAE can be reduced on the tested state-of-the-art forecasting models. The source code will be available at https://github.com/ergoucao/ERKG.
Authors: Zhuoxuan Liang, Wei Li, Dalin Zhang, Yidan Chen, Zhihong Wang, Xiangping Zheng, Moustafa Youssef
Abstract: With the rapid growth of the Internet of Things and Cyber-Physical Systems, widespread sensor deployment has become essential. However, the high costs of building sensor networks limit their scale and coverage, making fine-grained deployment challenging. Inductive Spatio-Temporal Kriging (ISK) addresses this issue by introducing virtual sensors. Based on graph neural networks (GNNs) extracting the relationships between physical and virtual sensors, ISK can infer the measurements of virtual sensors from physical sensors. However, current ISK methods rely on conventional message-passing mechanisms and network architectures, without effectively extracting spatio-temporal features of physical sensors and focusing on representing virtual sensors. Additionally, existing graph construction methods face issues of sparse and noisy connections, destroying ISK performance. To address these issues, we propose DarkFarseer, a novel ISK framework with three key components. First, we propose the Neighbor Hidden Style Enhancement module with a style transfer strategy to enhance the representation of virtual nodes in a temporal-then-spatial manner to better extract the spatial relationships between physical and virtual nodes. Second, we propose Virtual-Component Contrastive Learning, which aims to enrich the node representation by establishing the association between the patterns of virtual nodes and the regional patterns within graph components. Lastly, we design a Similarity-Based Graph Denoising Strategy, which reduces the connectivity strength of noisy connections around virtual nodes and their neighbors based on their temporal information and regional spatial patterns. Extensive experiments demonstrate that DarkFarseer significantly outperforms existing ISK methods.
Authors: Kavi Gupta, Kate Sanders, Armando Solar-Lezama
Abstract: Can LLMs pick up language structure from examples? Evidence in prior work seems to indicate yes, as pretrained models repeatedly demonstrate the ability to adapt to new language structures and vocabularies. However, this line of research typically considers languages that are present within common pretraining datasets, or otherwise share notable similarities with these seen languages. In contrast, in this work we attempt to measure models' language understanding capacity while circumventing the risk of dataset recall. We parameterize large families of language tasks recognized by deterministic finite automata (DFAs), and can thus sample novel language reasoning problems to fairly evaulate LLMs regardless of training data. We find that, even in the strikingly simple setting of 3-state DFAs, LLMs underperform unparameterized ngram models on both language recognition and synthesis tasks. These results suggest that LLMs struggle to match the ability of basic language models in recognizing and reasoning over languages that are sufficiently distinct from the ones they see at training time, underscoring the distinction between learning individual languages and possessing a general theory of language.
Authors: Arthur Aubret, C\'eline Teuli\`ere, Jochen Triesch
Abstract: Recent successes in self-supervised learning (SSL) model spatial co-occurrences of visual features either by masking portions of an image or by aggressively cropping it. Here, we propose a new way to model spatial co-occurrences by aligning local representations (before pooling) with a global image representation. We present CO-SSL, a family of instance discrimination methods and show that it outperforms previous methods on several datasets, including ImageNet-1K where it achieves 71.5% of Top-1 accuracy with 100 pre-training epochs. CO-SSL is also more robust to noise corruption, internal corruption, small adversarial attacks, and large training crop sizes. Our analysis further indicates that CO-SSL learns highly redundant local representations, which offers an explanation for its robustness. Overall, our work suggests that aligning local and global representations may be a powerful principle of unsupervised category learning.
Authors: Shayan Mohajer Hamidi, En-Hui Yang
Abstract: Inverse problems are prevalent across various disciplines in science and engineering. In the field of computer vision, tasks such as inpainting, deblurring, and super-resolution are commonly formulated as inverse problems. Recently, diffusion models (DMs) have emerged as a promising approach for addressing noisy linear inverse problems, offering effective solutions without requiring additional task-specific training. Specifically, with the prior provided by DMs, one can sample from the posterior by finding the likelihood. Since the likelihood is intractable, it is often approximated in the literature. However, this approximation compromises the quality of the generated images. To overcome this limitation and improve the effectiveness of DMs in solving inverse problems, we propose an information-theoretic approach. Specifically, we maximize the conditional mutual information $\mathrm{I}(\boldsymbol{x}_0; \boldsymbol{y} | \boldsymbol{x}_t)$, where $\boldsymbol{x}_0$ represents the reconstructed signal, $\boldsymbol{y}$ is the measurement, and $\boldsymbol{x}_t$ is the intermediate signal at stage $t$. This ensures that the intermediate signals $\boldsymbol{x}_t$ are generated in a way that the final reconstructed signal $\boldsymbol{x}_0$ retains as much information as possible about the measurement $\boldsymbol{y}$. We demonstrate that this method can be seamlessly integrated with recent approaches and, once incorporated, enhances their performance both qualitatively and quantitatively.
Authors: Shuangshuang He, Hongli Liang, Yuanting Zhang, Xingyuan Yuan
Abstract: High-resolution precipitation forecasts are crucial for providing accurate weather prediction and supporting effective responses to extreme weather events. Traditional numerical models struggle with stochastic subgrid-scale processes, while recent deep learning models often produce blurry results. To address these challenges, we propose a physics-inspired deep learning framework for high-resolution (0.05\textdegree{} $\times$ 0.05\textdegree{}) ensemble precipitation forecasting. Trained on ERA5 and CMPA high-resolution precipitation datasets, the framework integrates deterministic and probabilistic components. The deterministic model, based on a 3D SwinTransformer, captures average precipitation at mesoscale resolution and incorporates strategies to enhance performance, particularly for moderate to heavy rainfall. The probabilistic model employs conditional diffusion in latent space to account for uncertainties in residual precipitation at convective scales. During inference, ensemble members are generated by repeatedly sampling latent variables, enabling the model to represent precipitation uncertainty. Our model significantly enhances spatial resolution and forecast accuracy. Rank histogram shows that the ensemble system is reliable and unbiased. In a case study of heavy precipitation in southern China, the model outputs align more closely with observed precipitation distributions than ERA5, demonstrating superior capability in capturing extreme precipitation events. Additionally, 5-day real-time forecasts show good performance in terms of CSI scores.
Authors: Dravyansh Sharma, Arun Sai Suggala
Abstract: Classic algorithms for stochastic bandits typically use hyperparameters that govern their critical properties such as the trade-off between exploration and exploitation. Tuning these hyperparameters is a problem of great practical significance. However, this is a challenging problem and in certain cases is information theoretically impossible. To address this challenge, we consider a practically relevant transfer learning setting where one has access to offline data collected from several bandit problems (tasks) coming from an unknown distribution over the tasks. Our aim is to use this offline data to set the hyperparameters for a new task drawn from the unknown distribution. We provide bounds on the inter-task (number of tasks) and intra-task (number of arm pulls for each task) sample complexity for learning near-optimal hyperparameters on unseen tasks drawn from the distribution. Our results apply to several classic algorithms, including tuning the exploration parameters in UCB and LinUCB and the noise parameter in GP-UCB. Our experiments indicate the significance and effectiveness of the transfer of hyperparameters from offline problems in online learning with stochastic bandit feedback.
Authors: Charles O'Neill
Abstract: Self-attention mechanisms have revolutionised deep learning architectures, but their mathematical foundations remain incompletely understood. We establish that these mechanisms can be formalised through categorical algebra, presenting a framework that focuses on the linear components of self-attention. We prove that the query, key, and value maps in self-attention naturally form a parametric endofunctor in the 2-category $\mathbf{Para}(\mathbf{Vect})$ of parametric morphisms. We show that stacking multiple self-attention layers corresponds to constructing the free monad on this endofunctor. For positional encodings, we demonstrate that strictly additive position embeddings constitute monoid actions on the embedding space, while standard sinusoidal encodings, though not additive, possess a universal property among faithful position-preserving functors. We establish that the linear portions of self-attention exhibit natural equivariance properties with respect to permutations of input tokens. Finally, we prove that the ``circuits'' identified in mechanistic interpretability correspond precisely to compositions of parametric morphisms in our framework. This categorical perspective unifies geometric, algebraic, and interpretability-based approaches to transformer analysis, while making explicit the mathematical structures underlying attention mechanisms. Our treatment focuses exclusively on linear maps, setting aside nonlinearities like softmax and layer normalisation, which require more sophisticated categorical structures. Our results extend recent work on categorical foundations for deep learning while providing insights into the algebraic structure of attention mechanisms.
Authors: Shi Bin Hoo, Samuel M\"uller, David Salinas, Frank Hutter
Abstract: Foundation models have become popular in forecasting due to their ability to make accurate predictions, even with minimal fine-tuning on specific datasets. In this paper, we demonstrate how the newly released regression variant of TabPFN, a general tabular foundation model, can be applied to time series forecasting. We propose a straightforward approach, TabPFN-TS, which pairs TabPFN with simple feature engineering to achieve strong forecasting performance. Despite its simplicity and with only 11M parameters, TabPFN-TS outperforms Chronos-Mini, a model of similar size, and matches or even slightly outperforms Chronos-Large, which has 65-fold more parameters. A key strength of our method lies in its reliance solely on artificial data during pre-training, avoiding the need for large training datasets and eliminating the risk of benchmark contamination.
Authors: Stephan Goerttler, Yucheng Wang, Emadeldeen Eldele, Min Wu, Fei He
Abstract: Recent advancements in machine learning-based signal analysis, coupled with open data initiatives, have fuelled efforts in automatic sleep stage classification. Despite the proliferation of classification models, few have prioritised reducing model complexity, which is a crucial factor for practical applications. In this work, we introduce Multi-Scale and Attention Convolutional Neural Network (MSA-CNN), a lightweight architecture featuring as few as ~10,000 parameters. MSA-CNN leverages a novel multi-scale module employing complementary pooling to eliminate redundant filter parameters and dense convolutions. Model complexity is further reduced by separating temporal and spatial feature extraction and using cost-effective global spatial convolutions. This separation of tasks not only reduces model complexity but also mirrors the approach used by human experts in sleep stage scoring. We evaluated both small and large configurations of MSA-CNN against nine state-of-the-art baseline models across three public datasets, treating univariate and multivariate models separately. Our evaluation, based on repeated cross-validation and re-evaluation of all baseline models, demonstrated that the large MSA-CNN outperformed all baseline models on all three datasets in terms of accuracy and Cohen's kappa, despite its significantly reduced parameter count. Lastly, we explored various model variants and conducted an in-depth analysis of the key modules and techniques, providing deeper insights into the underlying mechanisms. The code for our models, baselines, and evaluation procedures is available at https://github.com/sgoerttler/MSA-CNN.
Authors: Ziyun Zou, Yinghui Jiang, Lian Shen, Juan Liu, Xiangrong Liu
Abstract: Spectral Graph Neural Networks effectively handle graphs with different homophily levels, with low-pass filter mining feature smoothness and high-pass filter capturing differences. When these distinct filters could naturally form two opposite views for self-supervised learning, the commonalities between the counterparts for the same node remain unexplored, leading to suboptimal performance. In this paper, a simple yet effective self-supervised contrastive framework, LOHA, is proposed to address this gap. LOHA optimally leverages low-pass and high-pass views by embracing "harmony in diversity". Rather than solely maximizing the difference between these distinct views, which may lead to feature separation, LOHA harmonizes the diversity by treating the propagation of graph signals from both views as a composite feature. Specifically, a novel high-dimensional feature named spectral signal trend is proposed to serve as the basis for the composite feature, which remains relatively unaffected by changing filters and focuses solely on original feature differences. LOHA achieves an average performance improvement of 2.8% over runner-up models on 9 real-world datasets with varying homophily levels. Notably, LOHA even surpasses fully-supervised models on several datasets, which underscores the potential of LOHA in advancing the efficacy of spectral GNNs for diverse graph structures.
Authors: Can Gao, Xiaofeng Tan, Jie Zhou, Weiping Ding, Witold Pedrycz
Abstract: Outlier detection refers to the identification of anomalous samples that deviate significantly from the distribution of normal data and has been extensively studied and used in a variety of practical tasks. However, most unsupervised outlier detection methods are carefully designed to detect specified outliers, while real-world data may be entangled with different types of outliers. In this study, we propose a fuzzy rough sets-based multi-scale outlier detection method to identify various types of outliers. Specifically, a novel fuzzy rough sets-based method that integrates relative fuzzy granule density is first introduced to improve the capability of detecting local outliers. Then, a multi-scale view generation method based on granular-ball computing is proposed to collaboratively identify group outliers at different levels of granularity. Moreover, reliable outliers and inliers determined by the three-way decision are used to train a weighted support vector machine to further improve the performance of outlier detection. The proposed method innovatively transforms unsupervised outlier detection into a semi-supervised classification problem and for the first time explores the fuzzy rough sets-based outlier detection from the perspective of multi-scale granular balls, allowing for high adaptability to different types of outliers. Extensive experiments carried out on both artificial and UCI datasets demonstrate that the proposed outlier detection method significantly outperforms the state-of-the-art methods, improving the results by at least 8.48% in terms of the Area Under the ROC Curve (AUROC) index. { The source codes are released at \url{https://github.com/Xiaofeng-Tan/MGBOD}. }
Authors: Anne Gagneux, Mathurin Massias, Emmanuel Soubies, R\'emi Gribonval
Abstract: Convex functions and their gradients play a critical role in mathematical imaging, from proximal optimization to Optimal Transport. The successes of deep learning has led many to use learning-based methods, where fixed functions or operators are replaced by learned neural networks. Regardless of their empirical superiority, establishing rigorous guarantees for these methods often requires to impose structural constraints on neural architectures, in particular convexity. The most popular way to do so is to use so-called Input Convex Neural Networks (ICNNs). In order to explore the expressivity of ICNNs, we provide necessary and sufficient conditions for a ReLU neural network to be convex. Such characterizations are based on product of weights and activations, and write nicely for any architecture in the path-lifting framework. As particular applications, we study our characterizations in depth for 1 and 2-hidden-layer neural networks: we show that every convex function implemented by a 1-hidden-layer ReLU network can be also expressed by an ICNN with the same architecture; however this property no longer holds with more layers. Finally, we provide a numerical procedure that allows an exact check of convexity for ReLU neural networks with a large number of affine regions.
Authors: Andreas Athanasopoulos, Anne-Marie George, Christos Dimitrakakis
Abstract: We consider a learning problem for the stable marriage model under unknown preferences for the left side of the market. We focus on the centralized case, where at each time step, an online platform matches the agents, and obtains a noisy evaluation reflecting their preferences. Our aim is to quickly identify the stable matching that is left-side optimal, rendering this a pure exploration problem with bandit feedback. We specifically aim to find Probably Correct Optimal Stable Matchings and present several bandit algorithms to do so. Our findings provide a foundational understanding of how to efficiently gather and utilize preference information to identify the optimal stable matching in two-sided markets under uncertainty. An experimental analysis on synthetic data complements theoretical results on sample complexities for the proposed methods.
Authors: Duygu Sezen Islakoglu, Jan-Christoph Kalo
Abstract: Large Language Models (LLMs) have achieved remarkable success in various NLP tasks, yet they still face significant challenges in reasoning and arithmetic. Temporal reasoning, a critical component of natural language understanding, has raised increasing research attention. However, comprehensive testing of Allen's interval relations (e.g., before, after, during) -- a fundamental framework for temporal relationships -- remains underexplored. To fill this gap, we present ChronoSense, a new benchmark for evaluating LLMs' temporal understanding. It includes 16 tasks, focusing on identifying the Allen relation between two temporal events and temporal arithmetic, using both abstract events and real-world data from Wikidata. We assess the performance of seven recent LLMs using this benchmark and the results indicate that models handle Allen relations, even symmetrical ones, quite differently. Moreover, the findings suggest that the models may rely on memorization to answer time-related questions. Overall, the models' low performance highlights the need for improved temporal understanding in LLMs and ChronoSense offers a robust framework for future research in this area. Our dataset and the source code are available at https://github.com/duyguislakoglu/chronosense.
Authors: Tianhua Chen
Abstract: This paper explores foundational and applied aspects of survival analysis, using fall risk assessment as a case study. It revisits key time-related probability distributions and statistical methods, including logistic regression, Poisson regression, Exponential regression, and the Cox Proportional Hazards model, offering a unified perspective on their relationships within the survival analysis framework. A contribution of this work is the step-by-step derivation and clarification of the relationships among these models, particularly demonstrating that Poisson regression in the survival context is a specific case of the Cox model. These insights address gaps in understanding and reinforce the simplicity and interpretability of survival models. The paper also emphasizes the practical utility of survival analysis by connecting theoretical insights with real-world applications. In the context of fall detection, it demonstrates how these models can simultaneously predict fall risk, analyze contributing factors, and estimate time-to-event outcomes within a single streamlined framework. In contrast, advanced deep learning methods often require complex post-hoc interpretation and separate training for different tasks particularly when working with structured numerical data. This highlights the enduring relevance of classical statistical frameworks and makes survival models especially valuable in healthcare settings, where explainability and robustness are critical. By unifying foundational concepts and offering a cohesive perspective on time-to-event analysis, this work serves as an accessible resource for understanding survival models and applying them effectively to diverse analytical challenges.
Authors: Th\'eophane Vallaeys, Matthew Muckley, Jakob Verbeek, Matthijs Douze
Abstract: Vector quantization is a fundamental technique for compression and large-scale nearest neighbor search. For high-accuracy operating points, multi-codebook quantization associates data vectors with one element from each of multiple codebooks. An example is residual quantization (RQ), which iteratively quantizes the residual error of previous steps. Dependencies between the different parts of the code are, however, ignored in RQ, which leads to suboptimal rate-distortion performance. QINCo recently addressed this inefficiency by using a neural network to determine the quantization codebook in RQ based on the vector reconstruction from previous steps. In this paper we introduce QINCo2 which extends and improves QINCo with (i) improved vector encoding using codeword pre-selection and beam-search, (ii) a fast approximate decoder leveraging codeword pairs to establish accurate short-lists for search, and (iii) an optimized training procedure and network architecture. We conduct experiments on four datasets to evaluate QINCo2 for vector compression and billion-scale nearest neighbor search. We obtain outstanding results in both settings, improving the state-of-the-art reconstruction MSE by 34% for 16-byte vector compression on BigANN, and search accuracy by 24% with 8-byte encodings on Deep1M.
Authors: Joshua Southern, Yam Eitan, Guy Bar-Shalom, Michael Bronstein, Haggai Maron, Fabrizio Frasca
Abstract: We propose an expressive and efficient approach that combines the strengths of two prominent extensions of Graph Neural Networks (GNNs): Subgraph GNNs and Structural Encodings (SEs). Our approach leverages walk-based centrality measures, both as a powerful form of SE and also as a subgraph selection strategy for Subgraph GNNs. By drawing a connection to perturbation analysis, we highlight the effectiveness of centrality-based sampling, and show it significantly reduces the computational burden associated with Subgraph GNNs. Further, we combine our efficient Subgraph GNN with SEs derived from the calculated centrality and demonstrate this hybrid approach, dubbed HyMN, gains in discriminative power. HyMN effectively addresses the expressiveness limitations of Message Passing Neural Networks (MPNNs) while mitigating the computational costs of Subgraph GNNs. Through a series of experiments on synthetic and real-world tasks, we show it outperforms other subgraph sampling approaches while being competitive with full-bag Subgraph GNNs and other state-of-the-art approaches with a notably reduced runtime.
Authors: Chao Feng, Yuanzhe Gao, Alberto Huertas Celdran, Gerome Bovet, Burkhard Stiller
Abstract: Federated Learning (FL) is widely recognized as a privacy-preserving machine learning paradigm due to its model-sharing mechanism that avoids direct data exchange. However, model training inevitably leaves exploitable traces that can be used to infer sensitive information. In Decentralized FL (DFL), the overlay topology significantly influences its models' convergence, robustness, and security. This study explores the feasibility of inferring the overlay topology of DFL systems based solely on model behavior, introducing a novel Topology Inference Attack. A taxonomy of topology inference attacks is proposed, categorizing them by the attacker's capabilities and knowledge. Practical attack strategies are developed for different scenarios, and quantitative experiments are conducted to identify key factors influencing the attack effectiveness. Experimental results demonstrate that analyzing only the public models of individual nodes can accurately infer the DFL topology, underscoring the risk of sensitive information leakage in DFL systems. This finding offers valuable insights for improving privacy preservation in decentralized learning environments.
Authors: Panagiotis Misiakos, Markus P\"uschel
Abstract: We introduce DAG-TFRC, a novel method for learning directed acyclic graphs (DAGs) from time series with few root causes. By this, we mean that the data are generated by a small number of events at certain, unknown nodes and time points under a structural vector autoregression model. For such data, we (i) learn the DAGs representing both the instantaneous and time-lagged dependencies between nodes, and (ii) discover the location and time of the root causes. For synthetic data with few root causes, DAG-TFRC shows superior performance in accuracy and runtime over prior work, scaling up to thousands of nodes. Experiments on simulated and real-world financial data demonstrate the viability of our sparse root cause assumption. On S&P 500 data, DAG-TFRC successfully clusters stocks by sectors and discovers major stock movements as root causes.
Authors: Zhihao Jia, Qi Pang, Trung Tran, David Woodruff, Zhihao Zhang, Wenting Zheng
Abstract: In this work, we study the experts problem in the distributed setting where an expert's cost needs to be aggregated across multiple servers. Our study considers various communication models such as the message-passing model and the broadcast model, along with multiple aggregation functions, such as summing and taking the $\ell_p$ norm of an expert's cost across servers. We propose the first communication-efficient protocols that achieve near-optimal regret in these settings, even against a strong adversary who can choose the inputs adaptively. Additionally, we give a conditional lower bound showing that the communication of our protocols is nearly optimal. Finally, we implement our protocols and demonstrate empirical savings on the HPO-B benchmarks.
Authors: Jing Zhang, Hui Gao, Peng Zhang, Shuzhen Sun, Chang Yang, Yuexian Hou
Abstract: LoRA (Low-Rank Adaptation) is a widely used model fine-tuning method. In fine-tuning, the law among model performance, model parameters, and data complexity has been a focal issue in the field. Existing methods often leverage external metrics (such as cross-entropy or perplexity) to evaluate model performance. In the fine-tuning process for large models, two types of knowledge are typically involved: the frozen, general knowledge acquired by the model during pre-training and the new knowledge learned through the LoRA module from the current data. Generally, the less LoRA's learned knowledge relies on the large model, the more it captures the specific knowledge of new data, thereby enhancing its adaptability to new tasks. However, external metrics do not readily capture the dependency relationship between these two types of knowledge. Therefore, we designed an internal metric based on the Mutual Information Upper Bound (MIUB) theory to investigate the scaling law of large-model LoRA fine-tuning. In our experiments, we validated this approach on benchmark datasets, using the Llama3-8B and Phi3-3B models. The results show that the proposed MIUB metric aligns more accurately and stably with the scaling law of LoRA fine-tuning compared to cross-entropy and perplexity.
Authors: Muyun Li, Aaron Fainman, Stefan Vlaski
Abstract: Decentralized learning strategies allow a collection of agents to learn efficiently from local data sets without the need for central aggregation or orchestration. Current decentralized learning paradigms typically rely on an averaging mechanism to encourage agreement in the parameter space. We argue that in the context of deep neural networks, which are often over-parameterized, encouraging consensus of the neural network outputs, as opposed to their parameters can be more appropriate. This motivates the development of a new decentralized learning algorithm, termed DRT diffusion, based on deep relative trust (DRT), a recently introduced similarity measure for neural networks. We provide convergence analysis for the proposed strategy, and numerically establish its benefit to generalization, especially with sparse topologies, in an image classification task.
Authors: Andrii Krutsylo
Abstract: We propose a scalable Forward-Forward (FF) algorithm that eliminates the need for backpropagation by training each layer separately. Unlike backpropagation, FF avoids backward gradients and can be more modular and memory efficient, making it appealing for large networks. We extend FF to modern convolutional architectures, such as MobileNetV3 and ResNet18, by introducing a new way to compute losses for convolutional layers. Experiments show that our method achieves performance comparable to standard backpropagation. Furthermore, when we divide the network into blocks, such as the residual blocks in ResNet, and apply backpropagation only within each block, but not across blocks, our hybrid design tends to outperform backpropagation baselines while maintaining a similar training speed. Finally, we present experiments on small datasets and transfer learning that confirm the adaptability of our method.
Authors: Andrew Chang, Viswadruth Akkaraju, Ray McFadden Cogliano, David Poeppel, Dustin Freeman
Abstract: Videoconferencing is now a frequent mode of communication in both professional and informal settings, yet it often lacks the fluidity and enjoyment of in-person conversation. This study leverages multimodal machine learning to predict moments of negative experience in videoconferencing. We sampled thousands of short clips from the RoomReader corpus, extracting audio embeddings, facial actions, and body motion features to train models for identifying low conversational fluidity, low enjoyment, and classifying conversational events (backchanneling, interruption, or gap). Our best models achieved an ROC-AUC of up to 0.87 on hold-out videoconference sessions, with domain-general audio features proving most critical. This work demonstrates that multimodal audio-video signals can effectively predict high-level subjective conversational outcomes. In addition, this is a contribution to research on videoconferencing user experience by showing that multimodal machine learning can be used to identify rare moments of negative user experience for further study or mitigation.
Authors: Sudeep Salgia, Nikola Pavlovic, Yuejie Chi, Qing Zhao
Abstract: We consider the problem of differentially private stochastic convex optimization (DP-SCO) in a distributed setting with $M$ clients, where each of them has a local dataset of $N$ i.i.d. data samples from an underlying data distribution. The objective is to design an algorithm to minimize a convex population loss using a collaborative effort across $M$ clients, while ensuring the privacy of the local datasets. In this work, we investigate the accuracy-communication-privacy trade-off for this problem. We establish matching converse and achievability results using a novel lower bound and a new algorithm for distributed DP-SCO based on Vaidya's plane cutting method. Thus, our results provide a complete characterization of the accuracy-communication-privacy trade-off for DP-SCO in the distributed setting.
Authors: Daria de Tinguy, Tim Verbelen, Bart Dhoedt
Abstract: Inspired by animal navigation strategies, we introduce a novel computational model to navigate and map a space rooted in biologically inspired principles. Animals exhibit extraordinary navigation prowess, harnessing memory, imagination, and strategic decision-making to traverse complex and aliased environments adeptly. Our model aims to replicate these capabilities by incorporating a dynamically expanding cognitive map over predicted poses within an Active Inference framework, enhancing our agent's generative model plasticity to novelty and environmental changes. Through structure learning and active inference navigation, our model demonstrates efficient exploration and exploitation, dynamically expanding its model capacity in response to anticipated novel un-visited locations and updating the map given new evidence contradicting previous beliefs. Comparative analyses in mini-grid environments with the Clone-Structured Cognitive Graph model (CSCG), which shares similar objectives, highlight our model's ability to rapidly learn environmental structures within a single episode, with minimal navigation overlap. Our model achieves this without prior knowledge of observation and world dimensions, underscoring its robustness and efficacy in navigating intricate environments.
Authors: Marco Minici, Luca Luceri, Francesco Fabbri, Emilio Ferrara
Abstract: Social media platforms have become vital spaces for public discourse, serving as modern agor\'as where a wide range of voices influence societal narratives. However, their open nature also makes them vulnerable to exploitation by malicious actors, including state-sponsored entities, who can conduct information operations (IOs) to manipulate public opinion. The spread of misinformation, false news, and misleading claims threatens democratic processes and societal cohesion, making it crucial to develop methods for the timely detection of inauthentic activity to protect the integrity of online discourse. In this work, we introduce a methodology designed to identify users orchestrating information operations, a.k.a. \textit{IO drivers}, across various influence campaigns. Our framework, named \texttt{IOHunter}, leverages the combined strengths of Language Models and Graph Neural Networks to improve generalization in \emph{supervised}, \emph{scarcely-supervised}, and \emph{cross-IO} contexts. Our approach achieves state-of-the-art performance across multiple sets of IOs originating from six countries, significantly surpassing existing approaches. This research marks a step toward developing Graph Foundation Models specifically tailored for the task of IO detection on social media platforms.
Authors: Oudom Hean, Utsha Saha, Binita Saha
Abstract: In recent years, Large Language Models (LLMs) have emerged as a transformative development in artificial intelligence (AI), drawing significant attention from industry and academia. Trained on vast datasets, these sophisticated AI systems exhibit impressive natural language processing and content generation capabilities. This paper explores the potential of LLMs to address key challenges in personal finance, focusing on the United States. We evaluate several leading LLMs, including OpenAI's ChatGPT, Google's Gemini, Anthropic's Claude, and Meta's Llama, to assess their effectiveness in providing accurate financial advice on topics such as mortgages, taxes, loans, and investments. Our findings show that while these models achieve an average accuracy rate of approximately 70%, they also display notable limitations in certain areas. Specifically, LLMs struggle to provide accurate responses for complex financial queries, with performance varying significantly across different topics. Despite these limitations, the analysis reveals notable improvements in newer versions of these models, highlighting their growing utility for individuals and financial advisors. As these AI systems continue to evolve, their potential for advancing AI-driven applications in personal finance becomes increasingly promising.
Authors: Jiahao Qin, Feng Liu
Abstract: Electrocardiogram (ECG) analysis plays a crucial role in diagnosing cardiovascular diseases, but accurate interpretation of these complex signals remains challenging. This paper introduces a novel multimodal framework(GAF-FusionNet) for ECG classification that integrates time-series analysis with image-based representation using Gramian Angular Fields (GAF). Our approach employs a dual-layer cross-channel split attention module to adaptively fuse temporal and spatial features, enabling nuanced integration of complementary information. We evaluate GAF-FusionNet on three diverse ECG datasets: ECG200, ECG5000, and the MIT-BIH Arrhythmia Database. Results demonstrate significant improvements over state-of-the-art methods, with our model achieving 94.5\%, 96.9\%, and 99.6\% accuracy on the respective datasets. Our code will soon be available at https://github.com/Cross-Innovation-Lab/GAF-FusionNet.git.
URLs: https://github.com/Cross-Innovation-Lab/GAF-FusionNet.git.
Authors: Alexander Kozachinskiy, Alexander Shen, Tomasz Steifer
Abstract: In perpetual voting, multiple decisions are made at different moments in time. Taking the history of previous decisions into account allows us to satisfy properties such as proportionality over periods of time. In this paper, we consider the following question: is there a perpetual approval voting method that guarantees that no voter is dissatisfied too many times? We identify a sufficient condition on voter behavior -- which we call 'bounded conflicts' condition -- under which a sublinear growth of dissatisfaction is possible. We provide a tight upper bound on the growth of dissatisfaction under bounded conflicts, using techniques from Kolmogorov complexity. We also observe that the approval voting with binary choices mimics the machine learning setting of prediction with expert advice. This allows us to present a voting method with sublinear guarantees on dissatisfaction under bounded conflicts, based on the standard techniques from prediction with expert advice.
Authors: Yuwei Du, Xinyue Liu, Wenxin Liang, Linlin Zong, Xianchao Zhang
Abstract: Temporal knowledge graph (TKG) reasoning has become a hot topic due to its great value in many practical tasks. The key to TKG reasoning is modeling the structural information and evolutional patterns of the TKGs. While great efforts have been devoted to TKG reasoning, the structural and evolutional characteristics of real-world networks have not been considered. In the aspect of structure, real-world networks usually exhibit clear community structure and scale-free (long-tailed distribution) properties. In the aspect of evolution, the impact of an event decays with the time elapsing. In this paper, we propose a novel TKG reasoning model called Hawkes process-based Evolutional Representation Learning Network (HERLN), which learns structural information and evolutional patterns of a TKG simultaneously, considering the characteristics of real-world networks: community structure, scale-free and temporal decaying. First, we find communities in the input TKG to make the encoding get more similar intra-community embeddings. Second, we design a Hawkes process-based relational graph convolutional network to cope with the event impact-decaying phenomenon. Third, we design a conditional decoding method to alleviate biases towards frequent entities caused by long-tailed distribution. Experimental results show that HERLN achieves significant improvements over the state-of-the-art models.
Authors: Mohammad Nadeem, Shahab Saquib Sohail, Erik Cambria, Bj\"orn W. Schuller, Amir Hussain
Abstract: The advent of text-to-video generation models has revolutionized content creation as it produces high-quality videos from textual prompts. However, concerns regarding inherent biases in such models have prompted scrutiny, particularly regarding gender representation. Our study investigates the presence of gender bias in OpenAI's Sora, a state-of-the-art text-to-video generation model. We uncover significant evidence of bias by analyzing the generated videos from a diverse set of gender-neutral and stereotypical prompts. The results indicate that Sora disproportionately associates specific genders with stereotypical behaviors and professions, which reflects societal prejudices embedded in its training data.
Authors: Alexander Du, Yingwu Zhu
Abstract: Estimating 6D object poses from RGB images is challenging because the lack of depth information requires inferring a three dimensional structure from 2D projections. Traditional methods often rely on deep learning with grid based data structures but struggle to capture complex dependencies among extracted features. To overcome this, we introduce a graph based representation derived directly from images, where spatial temporal features of each pixel serve as nodes, and relationships between them are defined through node connectivity and spatial interactions. We also employ feature selection mechanisms that use spatial attention and self attention distillation, along with a Legendre convolution layer leveraging the orthogonality of Legendre polynomials for numerical stability. Experiments on the LINEMOD, Occluded LINEMOD, and YCB Video datasets demonstrate that our method outperforms nine existing approaches and achieves state of the art benchmark in object pose estimation.
Authors: Sharvaree Vadgama, Mohammad Mohaiminul Islam, Domas Buracus, Christian Shewmake, Erik Bekkers
Abstract: This paper explores the key factors that influence the performance of models working with point clouds, across different tasks of varying geometric complexity. In this work, we explore the trade-offs between flexibility and weight-sharing introduced by equivariant layers, assessing when equivariance boosts or detracts from performance. It is often argued that providing more information as input improves a model's performance. However, if this additional information breaks certain properties, such as $\SE(3)$ equivariance, does it remain beneficial? We identify the key aspects of equivariant and non-equivariant architectures that drive success in different tasks by benchmarking them on segmentation, regression, and generation tasks across multiple datasets with increasing complexity. We observe a positive impact of equivariance, which becomes more pronounced with increasing task complexity, even when strict equivariance is not required.
Authors: Joao Fonseca, Andrew Bell, Julia Stoyanovich
Abstract: Large Language Models (LLMs) have been shown to be susceptible to jailbreak attacks, or adversarial attacks used to illicit high risk behavior from a model. Jailbreaks have been exploited by cybercriminals and blackhat actors to cause significant harm, highlighting the critical need to safeguard widely-deployed models. Safeguarding approaches, which include fine-tuning models or having LLMs "self-reflect", may lengthen the inference time of a model, incur a computational penalty, reduce the semantic fluency of an output, and restrict ``normal'' model behavior. Importantly, these Safety-Performance Trade-offs (SPTs) remain an understudied area. In this work, we introduce a novel safeguard, called SafeNudge, that combines Controlled Text Generation with "nudging", or using text interventions to change the behavior of a model. SafeNudge triggers during text-generation while a jailbreak attack is being executed, and can reduce successful jailbreak attempts by 30% by guiding the LLM towards a safe responses. It adds minimal latency to inference and has a negligible impact on the semantic fluency of outputs. Further, we allow for tunable SPTs. SafeNudge is open-source and available through https://pypi.org/, and is compatible with models loaded with the Hugging Face "transformers" library.
URLs: https://pypi.org/,
Authors: Elhoucine Elfatimi, Lahcen El fatimi
Abstract: Recent advancements in model checking have demonstrated significant potential across diverse applications, particularly in signal and image analysis. Medical imaging stands out as a critical domain where model checking can be effectively applied to design and evaluate robust frameworks. These frameworks facilitate automatic and semi-automatic delineation of regions of interest within images, aiding in accurate segmentation. This paper provides a comprehensive analysis of recent works leveraging spatial logic to develop operators and tools for identifying regions of interest, including tumorous and non-tumorous areas. Additionally, we examine the challenges inherent to spatial model-checking techniques, such as variability in ground truth data and the need for streamlined procedures suitable for routine clinical practice.
Authors: Kaleem Ullah Qasim, Jiashu Zhang, Tariq Alsahfi, Ateeq Ur Rehman Butt
Abstract: Enhancing the reasoning capabilities of Large Language Models remains a critical challenge in artificial intelligence. We introduce RDoLT, Recursive Decomposition of Logical Thought prompting, a novel framework that significantly boosts LLM reasoning performance. RDoLT is built on three key innovations: (1) recursively breaking down complex reasoning tasks into sub-tasks of progressive complexity; (2) employing an advanced selection and scoring mechanism to identify the most promising reasoning thoughts; and (3) integrating a knowledge propagation module that mimics human learning by keeping track of strong and weak thoughts for information propagation. Our approach was evaluated across multiple benchmarks, including GSM8K, SVAMP, MultiArith, LastLetterConcatenation, and Gaokao2023 Math. The results demonstrate that RDoLT consistently outperforms existing state-of-the-art techniques, achieving a 90.98 percent accuracy on GSM8K with ChatGPT-4, surpassing state-of-the-art techniques by 6.28 percent. Similar improvements were observed on other benchmarks, with accuracy gains ranging from 5.5 percent to 6.75 percent. These findings highlight RDoLT's potential to advance prompt engineering, offering a more effective and generalizable approach to complex reasoning tasks.
Authors: Petr Kisselev, Padmanabhan Seshaiyer
Abstract: Graph convolutional neural networks (GCNs) have shown tremendous promise in addressing data-intensive challenges in recent years. In particular, some attempts have been made to improve predictions of Susceptible-Infected-Recovered (SIR) models by incorporating human mobility between metapopulations and using graph approaches to estimate corresponding hyperparameters. Recently, researchers have found that a hybrid GCN-SIR approach outperformed existing methodologies when used on the data collected on a precinct level in Japan. In our work, we extend this approach to data collected from the continental US, adjusting for the differing mobility patterns and varying policy responses. We also develop the strategy for real-time continuous estimation of the reproduction number and study the accuracy of model predictions for the overall population as well as individual states. Strengths and limitations of the GCN-SIR approach are discussed as a potential candidate for modeling disease dynamics.
Authors: Ollie Liu, Sami Jaghouar, Johannes Hagemann, Shangshang Wang, Jason Wiemels, Jeff Kaufman, Willie Neiswanger
Abstract: We pretrain METAGENE-1, a 7-billion-parameter autoregressive transformer model, which we refer to as a metagenomic foundation model, on a novel corpus of diverse metagenomic DNA and RNA sequences comprising over 1.5 trillion base pairs. This dataset is sourced from a large collection of human wastewater samples, processed and sequenced using deep metagenomic (next-generation) sequencing methods. Unlike genomic models that focus on individual genomes or curated sets of specific species, the aim of METAGENE-1 is to capture the full distribution of genomic information present within this wastewater, to aid in tasks relevant to pandemic monitoring and pathogen detection. We carry out byte-pair encoding (BPE) tokenization on our dataset, tailored for metagenomic sequences, and then pretrain our model. In this paper, we first detail the pretraining dataset, tokenization strategy, and model architecture, highlighting the considerations and design choices that enable the effective modeling of metagenomic data. We then show results of pretraining this model on our metagenomic dataset, providing details about our losses, system metrics, and training stability over the course of pretraining. Finally, we demonstrate the performance of METAGENE-1, which achieves state-of-the-art results on a set of genomic benchmarks and new evaluations focused on human-pathogen detection and genomic sequence embedding, showcasing its potential for public health applications in pandemic monitoring, biosurveillance, and early detection of emerging health threats.
Authors: Akshay Balsubramani
Abstract: We develop the laws of thermodynamics in terms of general exponential families. By casting learning (log-loss minimization) problems in max-entropy and statistical mechanics terms, we translate thermodynamics results to learning scenarios. We extend the well-known way in which exponential families characterize thermodynamic and learning equilibria. Basic ideas of work and heat, and advanced concepts of thermodynamic cycles and equipartition of energy, find exact and useful counterparts in AI / statistics terms. These ideas have broad implications for quantifying and addressing distribution shift.
Authors: Joanna Bieri, Giorgi Butbaia, Edgar Costa, Alyson Deines, Kyu-Hwan Lee, David Lowry-Duda, Thomas Oliver, Yidi Qi, Tamara Veenstra
Abstract: In this paper, we conduct a data-scientific investigation of Maass forms. We find that averaging the Fourier coefficients of Maass forms with the same Fricke sign reveals patterns analogous to the recently discovered "murmuration" phenomenon, and that these patterns become more pronounced when parity is incorporated as an additional feature. Approximately 43% of the forms in our dataset have an unknown Fricke sign. For the remaining forms, we employ Linear Discriminant Analysis (LDA) to machine learn their Fricke sign, achieving 96% (resp. 94%) accuracy for forms with even (resp. odd) parity. We apply the trained LDA model to forms with unknown Fricke signs to make predictions. The average values based on the predicted Fricke signs are computed and compared to those for forms with known signs to verify the reasonableness of the predictions. Additionally, a subset of these predictions is evaluated against heuristic guesses provided by Hejhal's algorithm, showing a match approximately 95% of the time. We also use neural networks to obtain results comparable to those from the LDA model.
Authors: Renichiro Haba, Masayuki Ohzeki, Kazuyuki Tanaka
Abstract: Quantum annealing has garnered significant attention as meta-heuristics inspired by quantum physics for combinatorial optimization problems. Among its many applications, nonnegative/binary matrix factorization stands out for its complexity and relevance in unsupervised machine learning. The use of reverse annealing, a derivative procedure of quantum annealing to prioritize the search in a vicinity under a given initial state, helps improve its optimization performance in matrix factorization. This study proposes an improved strategy that integrates reverse annealing with a linear programming relaxation technique. Using relaxed solutions as the initial configuration for reverse annealing, we demonstrate improvements in optimization performance comparable to the exact optimization methods. Our experiments on facial image datasets show that our method provides better convergence than known reverse annealing methods. Furthermore, we investigate the effectiveness of relaxation-based initialization methods on randomized datasets, demonstrating a relationship between the relaxed solution and the optimal solution. This research underscores the potential of combining reverse annealing and classical optimization strategies to enhance optimization performance.
Authors: Zhaobin Mo, Yunlong Li, Xuan Di
Abstract: Safety-critical driving data is crucial for developing safe and trustworthy self-driving algorithms. Due to the scarcity of safety-critical data in naturalistic datasets, current approaches primarily utilize simulated or artificially generated images. However, there remains a gap in authenticity between these generated images and naturalistic ones. We propose a novel framework to augment the safety-critical driving data from the naturalistic dataset to address this issue. In this framework, we first detect vehicles using YOLOv5, followed by depth estimation and 3D transformation to simulate vehicle proximity and critical driving scenarios better. This allows for targeted modification of vehicle dynamics data to reflect potentially hazardous situations. Compared to the simulated or artificially generated data, our augmentation methods can generate safety-critical driving data with minimal compromise on image authenticity. Experiments using KITTI datasets demonstrate that a downstream self-driving algorithm trained on this augmented dataset performs superiorly compared to the baselines, which include SMOGN and importance sampling.
Authors: Umesh Yadav, Suman Niraula, Gaurav Kumar Gupta, Bicky Yadav
Abstract: This paper investigates the resilience of a ResNet-50 image classification model under two prominent security threats: Fast Gradient Sign Method (FGSM) adversarial attacks and malicious payload injection. Initially, the model attains a 53.33% accuracy on clean images. When subjected to FGSM perturbations, its overall accuracy remains unchanged; however, the model's confidence in incorrect predictions notably increases. Concurrently, a payload injection scheme is successfully executed in 93.33% of the tested samples, revealing how stealthy attacks can manipulate model predictions without degrading visual quality. These findings underscore the vulnerability of even high-performing neural networks and highlight the urgency of developing more robust defense mechanisms for security-critical applications.
Authors: Huixue Zhou, Hengrui Gu, Xi Liu, Kaixiong Zhou, Mingfu Liang, Yongkang Xiao, Srinivas Govindan, Piyush Chawla, Jiyan Yang, Xiangfei Meng, Huayu Li, Buyun Zhang, Liang Luo, Wen-Yen Chen, Yiping Han, Bo Long, Rui Zhang, Tianlong Chen
Abstract: The deployment of Large Language Models (LLMs) in recommender systems for predicting Click-Through Rates (CTR) necessitates a delicate balance between computational efficiency and predictive accuracy. This paper presents an optimization framework that combines Retrieval-Augmented Generation (RAG) with an innovative multi-head early exit architecture to concurrently enhance both aspects. By integrating Graph Convolutional Networks (GCNs) as efficient retrieval mechanisms, we are able to significantly reduce data retrieval times while maintaining high model performance. The early exit strategy employed allows for dynamic termination of model inference, utilizing real-time predictive confidence assessments across multiple heads. This not only quickens the responsiveness of LLMs but also upholds or improves their accuracy, making it ideal for real-time application scenarios. Our experiments demonstrate how this architecture effectively decreases computation time without sacrificing the accuracy needed for reliable recommendation delivery, establishing a new standard for efficient, real-time LLM deployment in commercial systems.
Authors: Wei Wu, Zizhen Deng, Chi Zhang, Can Liao, Jinzhuo Wang
Abstract: Addressing the unavoidable bias inherent in supervised aging clocks, we introduce Sundial, a novel framework that models molecular dynamics through a diffusion field, capturing both the population-level aging process and the individual-level relative aging order. Sundial enables unbiasedestimation of biological age and the forecast of aging roadmap. Fasteraging individuals from Sundial exhibit a higher disease risk compared to those identified from supervised aging clocks. This framework opens new avenues for exploring key topics, including age- and sex-specific aging dynamics and faster yet healthy aging paths.
Authors: Yaodan Xu, Sheng Zhou, Zhisheng Niu
Abstract: For servers incorporating parallel computing resources, batching is a pivotal technique for providing efficient and economical services at scale. Parallel computing resources exhibit heightened computational and energy efficiency when operating with larger batch sizes. However, in the realm of online services, the adoption of a larger batch size may lead to longer response times. This paper aims to provide a dynamic batching scheme that delicately balances latency and efficiency. The system is modeled as a batch service queue with size-dependent service times. Then, the design of dynamic batching is formulated as a semi-Markov decision process (SMDP) problem, with the objective of minimizing the weighted sum of average response time and average power consumption. A method is proposed to derive an approximate optimal SMDP solution, representing the chosen dynamic batching policy. By introducing an abstract cost to reflect the impact of "tail" states, the space complexity and the time complexity of the procedure can decrease by 63.5% and 98%, respectively. Numerical results showcase the superiority of SMDP-based batching policies across various parameter setups. Additionally, the proposed scheme exhibits noteworthy flexibility in balancing power consumption and latency.
Authors: Zongxia Li, Xiyang Wu, Hongyang Du, Huy Nghiem, Guangyao Shi
Abstract: Multimodal Vision Language Models (VLMs) have emerged as a transformative technology at the intersection of computer vision and natural language processing, enabling machines to perceive and reason about the world through both visual and textual modalities. For example, models such as CLIP, Claude, and GPT-4V demonstrate strong reasoning and understanding abilities on visual and textual data and beat classical single modality vision models on zero-shot classification. Despite their rapid advancements in research and growing popularity in applications, a comprehensive survey of existing studies on VLMs is notably lacking, particularly for researchers aiming to leverage VLMs in their specific domains. To this end, we provide a systematic overview of VLMs in the following aspects: model information of the major VLMs developed over the past five years (2019-2024); the main architectures and training methods of these VLMs; summary and categorization of the popular benchmarks and evaluation metrics of VLMs; the applications of VLMs including embodied agents, robotics, and video generation; the challenges and issues faced by current VLMs such as hallucination, fairness, and safety. Detailed collections including papers and model repository links are listed in https://github.com/zli12321/Awesome-VLM-Papers-And-Models.git.
URLs: https://github.com/zli12321/Awesome-VLM-Papers-And-Models.git.
Authors: Jianmin Chen, Kun Chen
Abstract: The generalized lasso is a natural generalization of the celebrated lasso approach to handle structural regularization problems. Many important methods and applications fall into this framework, including fused lasso, clustered lasso, and constrained lasso. To elevate its effectiveness in large-scale problems, extensive research has been conducted on the computational strategies of generalized lasso. However, to our knowledge, most studies are under the linear setup, with limited advances in non-Gaussian and non-linear models. We propose a majorization-minimization dual stagewise (MM-DUST) algorithm to efficiently trace out the full solution paths of the generalized lasso problem. The majorization technique is incorporated to handle different convex loss functions through their quadratic majorizers. Utilizing the connection between primal and dual problems and the idea of ``slow-brewing'' from stagewise learning, the minimization step is carried out in the dual space through a sequence of simple coordinate-wise updates on the dual coefficients with a small step size. Consequently, selecting an appropriate step size enables a trade-off between statistical accuracy and computational efficiency. We analyze the computational complexity of MM-DUST and establish the uniform convergence of the approximated solution paths. Extensive simulation studies and applications with regularized logistic regression and Cox model demonstrate the effectiveness of the proposed approach.
Authors: Chao Wang, Licheng Jiao, Jiaxuan Zhao, Lingling Li, Fang Liu, Shuyuan Yang
Abstract: Evolutionary algorithms (EAs) maintain populations through evolutionary operators to discover diverse solutions for complex tasks while gathering valuable knowledge, such as historical population data and fitness evaluations. However, traditional EAs face challenges in dynamically adapting to expanding knowledge bases, hindering the efficient exploitation of accumulated information and limiting adaptability to new situations. To address these issues, we introduce an Optimization Knowledge Adaptation Evolutionary Model (OKAEM), which features dynamic parameter adjustment using accumulated knowledge to enhance its optimization capabilities. OKAEM employs attention mechanisms to model the interactions among individuals, fitness landscapes, and genetic components separately, thereby parameterizing the evolutionary operators of selection, crossover, and mutation. These powerful learnable operators enable OKAEM to benefit from pre-learned extensive prior knowledge and self-tune with real-time evolutionary insights. Experimental results demonstrate that OKAEM: 1) exploits prior knowledge for significant performance gains across various knowledge transfer settings; 2) achieves competitive performance through self-tuning alone, even without prior knowledge; 3) outperforms state-of-the-art black-box baselines in a vision-language model tuning case; 4) can improve its optimization capabilities with growing knowledge; 5) is capable of emulating principles of natural selection and genetic recombination.
Authors: Tong Deng, Tianming Wang
Abstract: We consider the robust multi-dimensional scaling (RMDS) problem in this paper. The goal is to localize point locations from pairwise distances that may be corrupted by outliers. Inspired by classic MDS theories, and nonconvex works for the robust principal component analysis (RPCA) problem, we propose an alternating projection based algorithm that is further accelerated by the tangent space projection technique. For the proposed algorithm, if the outliers are sparse enough, we can establish linear convergence of the reconstructed points to the original points after centering and rotation alignment. Numerical experiments verify the state-of-the-art performances of the proposed algorithm.
Authors: Messi H. J. Lee
Abstract: Vision-Language Models trained on massive collections of human-generated data often reproduce and amplify societal stereotypes. One critical form of stereotyping reproduced by these models is homogeneity bias-the tendency to represent certain groups as more homogeneous than others. We investigate how this bias responds to hyperparameter adjustments in GPT-4, specifically examining sampling temperature and top p which control the randomness of model outputs. By generating stories about individuals from different racial and gender groups and comparing their similarities using vector representations, we assess both bias robustness and its relationship with hyperparameter values. We find that (1) homogeneity bias persists across most hyperparameter configurations, with Black Americans and women being represented more homogeneously than White Americans and men, (2) the relationship between hyperparameters and group representations shows unexpected non-linear patterns, particularly at extreme values, and (3) hyperparameter adjustments affect racial and gender homogeneity bias differently-while increasing temperature or decreasing top p can reduce racial homogeneity bias, these changes show different effects on gender homogeneity bias. Our findings suggest that while hyperparameter tuning may mitigate certain biases to some extent, it cannot serve as a universal solution for addressing homogeneity bias across different social group dimensions.
Authors: Kanefumi Matsuyama, Kefan Su, Jiangxing Wang, Deheng Ye, Zongqing Lu
Abstract: Cooperative multi-agent reinforcement learning (MARL) aims to develop agents that can collaborate effectively. However, most cooperative MARL methods overfit training agents, making learned policies not generalize well to unseen collaborators, which is a critical issue for real-world deployment. Some methods attempt to address the generalization problem but require prior knowledge or predefined policies of new teammates, limiting real-world applications. To this end, we propose a hierarchical MARL approach to enable generalizable cooperation via role diversity, namely CORD. CORD's high-level controller assigns roles to low-level agents by maximizing the role entropy with constraints. We show this constrained objective can be decomposed into causal influence in role that enables reasonable role assignment, and role heterogeneity that yields coherent, non-redundant role clusters. Evaluated on a variety of cooperative multi-agent tasks, CORD achieves better performance than baselines, especially in generalization tests. Ablation studies further demonstrate the efficacy of the constrained objective in generalizable cooperation.
Authors: Victor Nascimento Ribeiro, Nina S. T. Hirata
Abstract: Automatic License Plate Recognition (ALPR) involves extracting vehicle license plate information from image or a video capture. These systems have gained popularity due to the wide availability of low-cost surveillance cameras and advances in Deep Learning. Typically, video-based ALPR systems rely on multiple frames to detect the vehicle and recognize the license plates. Therefore, we propose a system capable of extracting exactly one frame per vehicle and recognizing its license plate characters from this singular image using an Optical Character Recognition (OCR) model. Early experiments show that this methodology is viable.
Authors: Marta Gentiloni-Silveri, Antonio Ocello
Abstract: Score-based Generative Models (SGMs) aim to sample from a target distribution by learning score functions using samples perturbed by Gaussian noise. Existing convergence bounds for SGMs in the $\mathcal{W}_2$-distance rely on stringent assumptions about the data distribution. In this work, we present a novel framework for analyzing $\mathcal{W}_2$-convergence in SGMs, significantly relaxing traditional assumptions such as log-concavity and score regularity. Leveraging the regularization properties of the Ornstein-Uhlenbeck (OU) process, we show that weak log-concavity of the data distribution evolves into log-concavity over time. This transition is rigorously quantified through a PDE-based analysis of the Hamilton-Jacobi-Bellman equation governing the log-density of the forward process. Moreover, we establish that the drift of the time-reversed OU process alternates between contractive and non-contractive regimes, reflecting the dynamics of concavity. Our approach circumvents the need for stringent regularity conditions on the score function and its estimators, relying instead on milder, more practical assumptions. We demonstrate the wide applicability of this framework through explicit computations on Gaussian mixture models, illustrating its versatility and potential for broader classes of data distributions.
Authors: Debjany Ghosh Aronno, Sumaiya Saeha
Abstract: Diabetic Retinopathy (DR) is a major cause of blindness worldwide, caused by damage to the blood vessels in the retina due to diabetes. Early detection and classification of DR are crucial for timely intervention and preventing vision loss. This work proposes an automated system for DR detection using Convolutional Neural Networks (CNNs) with a residual block architecture, which enhances feature extraction and model performance. To further improve the model's robustness, we incorporate advanced data augmentation techniques, specifically leveraging a Deep Convolutional Generative Adversarial Network (DCGAN) for generating diverse retinal images. This approach increases the variability of training data, making the model more generalizable and capable of handling real-world variations in retinal images. The system is designed to classify retinal images into five distinct categories, from No DR to Proliferative DR, providing an efficient and scalable solution for early diagnosis and monitoring of DR progression. The proposed model aims to support healthcare professionals in large-scale DR screening, especially in resource-constrained settings.
Authors: Mikhail Zotov (for the JEM-EUSO Collaboration), Pavel Zakharov (for the JEM-EUSO Collaboration)
Abstract: Fluorescence telescopes are among the key instruments used for studying ultra-high energy cosmic rays in all modern experiments. We use model data for a small ground-based telescope EUSO-TA to try some methods of machine learning and neural networks for recognizing tracks of extensive air showers in its data and for reconstruction of energy and arrival directions of primary particles. We also comment on the opportunities to use this approach for other fluorescence telescopes and outline possible ways of improving the performance of the suggested methods.
Authors: Ali Bavafa, Gholam-Ali Hossein-Zadeh
Abstract: Understanding the causal interactions in simple brain tasks, such as face detection, remains a challenging and ambiguous process for researchers. In this study, we address this issue by employing a novel causal discovery method -- Directed Acyclic Graphs via M-matrices for Acyclicity (DAGMA) -- to investigate the causal structure of the brain's face-selective network and gain deeper insights into its mechanism. Using natural movie stimuli, we extract causal network of face-selective regions and analyze how frames containing faces influence this network. Our findings reveal that the presence of faces in the stimuli have causal effect both on the number and strength of causal connections within the network. Additionally, our results highlight the crucial role of subcortical regions in satisfying causal sufficiency, emphasizing its importance in causal studies of brain. This study provides a new perspective on understanding the causal architecture of the face-selective network of the brain, motivating further research on neural causality.
Authors: Ali Ghanbarzade, Hossein Soleimani
Abstract: The increasing reliance on Global Navigation Satellite Systems (GNSS), particularly the Global Positioning System (GPS), underscores the urgent need to safeguard these technologies against malicious threats such as spoofing and jamming. As the backbone for positioning, navigation, and timing (PNT) across various applications including transportation, telecommunications, and emergency services GNSS is vulnerable to deliberate interference that poses significant risks. Spoofing attacks, which involve transmitting counterfeit GNSS signals to mislead receivers into calculating incorrect positions, can result in serious consequences, from navigational errors in civilian aviation to security breaches in military operations. Furthermore, the lack of inherent security measures within GNSS systems makes them attractive targets for adversaries. While GNSS/GPS jamming and spoofing systems consist of numerous components, the ability to distinguish authentic signals from malicious ones is essential for maintaining system integrity. Recent advancements in machine learning and deep learning provide promising avenues for enhancing detection and mitigation strategies against these threats. This paper addresses both spoofing and jamming by tackling real-world challenges through machine learning, deep learning, and computer vision techniques. Through extensive experiments on two real-world datasets related to spoofing and jamming detection using advanced algorithms, we achieved state of the art results. In the GNSS/GPS jamming detection task, we attained approximately 99% accuracy, improving performance by around 5% compared to previous studies. Additionally, we addressed a challenging tasks related to spoofing detection, yielding results that underscore the potential of machine learning and deep learning in this domain.
Authors: Tara Radvand, Mojtaba Abdolmaleki, Mohamed Mostagir, Ambuj Tewari
Abstract: Verifying the provenance of content is crucial to the function of many organizations, e.g., educational institutions, social media platforms, firms, etc. This problem is becoming increasingly difficult as text generated by Large Language Models (LLMs) becomes almost indistinguishable from human-generated content. In addition, many institutions utilize in-house LLMs and want to ensure that external, non-sanctioned LLMs do not produce content within the institution. In this paper, we answer the following question: Given a piece of text, can we identify whether it was produced by LLM $A$ or $B$ (where $B$ can be a human)? We model LLM-generated text as a sequential stochastic process with complete dependence on history and design zero-shot statistical tests to distinguish between (i) the text generated by two different sets of LLMs $A$ (in-house) and $B$ (non-sanctioned) and also (ii) LLM-generated and human-generated texts. We prove that the type I and type II errors for our tests decrease exponentially in the text length. In designing our tests, we derive concentration inequalities on the difference between log-perplexity and the average entropy of the string under $A$. Specifically, for a given string, we demonstrate that if the string is generated by $A$, the log-perplexity of the string under $A$ converges to the average entropy of the string under $A$, except with an exponentially small probability in string length. We also show that if $B$ generates the text, except with an exponentially small probability in string length, the log-perplexity of the string under $A$ converges to the average cross-entropy of $B$ and $A$. Lastly, we present preliminary experimental results to support our theoretical results. By enabling guaranteed (with high probability) finding of the origin of harmful LLM-generated text with arbitrary size, we can help fight misinformation.
Authors: Antoine Boutet, Zakaria El Kazdam, Lucas Magnana, Helain Zimmermann
Abstract: Rapid advances in Natural Language Processing (NLP) have revolutionized many fields, including healthcare. However, these advances raise significant privacy concerns, especially when models specialized on sensitive data can memorize and then expose and regurgitate confidential information. This paper presents a privacy-by-design language modeling approach to address the problem of language models anonymization, and thus promote their sharing. Specifically, we propose both a Masking Language Modeling (MLM) methodology to specialize a BERT-like language model, and a Causal Language Modeling (CLM) methodology to specialize a GPT-like model that avoids the model from memorizing direct and indirect identifying information present in the training data. We have comprehensively evaluated our approaches using medical datasets and compared them against different baselines. Our results indicate that by avoiding memorizing both direct and indirect identifiers during model specialization, our masking and causal language modeling schemes offer the best tradeoff for maintaining high privacy while retaining high utility.
Authors: Hongzhe Zhang, Arnab Auddy, Hongzhe Lee
Abstract: Linear discriminant analysis is a widely used method for classification. However, the high dimensionality of predictors combined with small sample sizes often results in large classification errors. To address this challenge, it is crucial to leverage data from related source models to enhance the classification performance of a target model. We propose to address this problem in the framework of transfer learning. In this paper, we present novel transfer learning methods via regularized random-effects linear discriminant analysis, where the discriminant direction is estimated as a weighted combination of ridge estimates obtained from both the target and source models. Multiple strategies for determining these weights are introduced and evaluated, including one that minimizes the estimation risk of the discriminant vector and another that minimizes the classification error. Utilizing results from random matrix theory, we explicitly derive the asymptotic values of these weights and the associated classification error rates in the high-dimensional setting, where $p/n \rightarrow \infty$, with $p$ representing the predictor dimension and $n$ the sample size. We also provide geometric interpretations of various weights and a guidance on which weights to choose. Extensive numerical studies, including simulations and analysis of proteomics-based 10-year cardiovascular disease risk classification, demonstrate the effectiveness of the proposed approach.
Authors: Bingjie Lu (Central South University), Han-Cheng Dan (Central South University), Yichen Zhang (Central South University), Zhetao Huang (Central South University)
Abstract: Mean texture depth (MTD) is pivotal in assessing the skid resistance of asphalt pavements and ensuring road safety. This study focuses on developing an automated system for extracting texture features and evaluating MTD based on pavement images. The contributions of this work are threefold: firstly, it proposes an economical method to acquire three-dimensional (3D) pavement texture data; secondly, it enhances 3D image processing techniques and formulates features that represent various aspects of texture; thirdly, it establishes multivariate prediction models that link these features with MTD values. Validation results demonstrate that the Gradient Boosting Tree (GBT) model achieves remarkable prediction stability and accuracy (R2 = 0.9858), and field tests indicate the superiority of the proposed method over other techniques, with relative errors below 10%. This method offers a comprehensive end-to-end solution for pavement quality evaluation, from images input to MTD predictions output.
Authors: Yinpeng Cai, Lexin Li, Linjun Zhang
Abstract: Large Language Models (LLMs) are rapidly gaining enormous popularity in recent years. However, the training of LLMs has raised significant privacy and legal concerns, particularly regarding the inclusion of copyrighted materials in their training data without proper attribution or licensing, which falls under the broader issue of data misappropriation. In this article, we focus on a specific problem of data misappropriation detection, namely, to determine whether a given LLM has incorporated data generated by another LLM. To address this issue, we propose embedding watermarks into the copyrighted training data and formulating the detection of data misappropriation as a hypothesis testing problem. We develop a general statistical testing framework, construct a pivotal statistic, determine the optimal rejection threshold, and explicitly control the type I and type II errors. Furthermore, we establish the asymptotic optimality properties of the proposed tests, and demonstrate its empirical effectiveness through intensive numerical experiments.
Authors: Avni Mittal, John Kalkhof, Anirban Mukhopadhyay, Arnav Bhavsar
Abstract: Denoising Diffusion Models (DDMs) are widely used for high-quality image generation and medical image segmentation but often rely on Unet-based architectures, leading to high computational overhead, especially with high-resolution images. This work proposes three NCA-based improvements for diffusion-based medical image segmentation. First, Multi-MedSegDiffNCA uses a multilevel NCA framework to refine rough noise estimates generated by lower level NCA models. Second, CBAM-MedSegDiffNCA incorporates channel and spatial attention for improved segmentation. Third, MultiCBAM-MedSegDiffNCA combines these methods with a new RGB channel loss for semantic guidance. Evaluations on Lesion segmentation show that MultiCBAM-MedSegDiffNCA matches Unet-based model performance with dice score of 87.84% while using 60-110 times fewer parameters, offering a more efficient solution for low resource medical settings.
Authors: Haifeng Jia, Xinyi Chen, Yichen Wei, Yifei Sun, Yibo Pi
Abstract: Ray tracing is widely employed to model the propagation of radio-frequency (RF) signal in complex environment. The modelling performance greatly depends on how accurately the target scene can be depicted, including the scene geometry and surface material properties. The advances in computer vision and LiDAR make scene geometry estimation increasingly accurate, but there still lacks scalable and efficient approaches to estimate the material reflectivity in real-world environment. In this work, we tackle this problem by learning the material reflectivity efficiently from the path loss of the RF signal from the transmitters to receivers. Specifically, we want the learned material reflection coefficients to minimize the gap between the predicted and measured powers of the receivers. We achieve this by translating the neural reflectance field from optics to RF domain by modelling both the amplitude and phase of RF signals to account for the multipath effects. We further propose a differentiable RF ray tracing framework that optimizes the neural reflectance field to match the signal strength measurements. We simulate a complex real-world environment for experiments and our simulation results show that the neural reflectance field can successfully learn the reflection coefficients for all incident angles. As a result, our approach achieves better accuracy in predicting the powers of receivers with significantly less training data compared to existing approaches.
Authors: No\'e Dia, M. J. Yantovski-Barth, Alexandre Adam, Micah Bowles, Laurence Perreault-Levasseur, Yashar Hezaveh, Anna Scaife
Abstract: Inferring sky surface brightness distributions from noisy interferometric data in a principled statistical framework has been a key challenge in radio astronomy. In this work, we introduce Imaging for Radio Interferometry with Score-based models (IRIS). We use score-based models trained on optical images of galaxies as an expressive prior in combination with a Gaussian likelihood in the uv-space to infer images of protoplanetary disks from visibility data of the DSHARP survey conducted by ALMA. We demonstrate the advantages of this framework compared with traditional radio interferometry imaging algorithms, showing that it produces plausible posterior samples despite the use of a misspecified galaxy prior. Through coverage testing on simulations, we empirically evaluate the accuracy of this approach to generate calibrated posterior samples.
Authors: Chao Liang, Linchao Zhu, Zongxin Yang, Wei Chen, Yi Yang
Abstract: We focus on the challenging problem of learning an unbiased classifier from a large number of potentially relevant but noisily labeled web images given only a few clean labeled images. This problem is particularly practical because it reduces the expensive annotation costs by utilizing freely accessible web images with noisy labels. Typically, prototypes are representative images or features used to classify or identify other images. However, in the few clean and many noisy scenarios, the class prototype can be severely biased due to the presence of irrelevant noisy images. The resulting prototypes are less compact and discriminative, as previous methods do not take into account the diverse range of images in the noisy web image collections. On the other hand, the relation modeling between noisy and clean images is not learned for the class prototype generation in an end-to-end manner, which results in a suboptimal class prototype. In this article, we introduce a similarity maximization loss named SimNoiPro. Our SimNoiPro first generates noise-tolerant hybrid prototypes composed of clean and noise-tolerant prototypes and then pulls them closer to each other. Our approach considers the diversity of noisy images by explicit division and overcomes the optimization discrepancy issue. This enables better relation modeling between clean and noisy images and helps extract judicious information from the noisy image set. The evaluation results on two extended few-shot classification benchmarks confirm that our SimNoiPro outperforms prior methods in measuring image relations and cleaning noisy data.
Authors: Marzieh Esnaashari, Nima Moradi
Abstract: In an era of escalating cyber threats, malware poses significant risks to individuals and organizations, potentially leading to data breaches, system failures, and substantial financial losses. This study addresses the urgent need for effective malware detection strategies by leveraging Machine Learning (ML) techniques on extensive datasets collected from Microsoft Windows Defender. Our research aims to develop an advanced ML model that accurately predicts malware vulnerabilities based on the specific conditions of individual machines. Moving beyond traditional signature-based detection methods, we incorporate historical data and innovative feature engineering to enhance detection capabilities. This study makes several contributions: first, it advances existing malware detection techniques by employing sophisticated ML algorithms; second, it utilizes a large-scale, real-world dataset to ensure the applicability of findings; third, it highlights the importance of feature analysis in identifying key indicators of malware infections; and fourth, it proposes models that can be adapted for enterprise environments, offering a proactive approach to safeguarding extensive networks against emerging threats. We aim to improve cybersecurity resilience, providing critical insights for practitioners in the field and addressing the evolving challenges posed by malware in a digital landscape. Finally, discussions on results, insights, and conclusions are presented.
Authors: Yixin Ji, Juntao Li, Hai Ye, Kaixin Wu, Jia Xu, Linjian Mo, Min Zhang
Abstract: The remarkable performance of the o1 model in complex reasoning demonstrates that test-time computing scaling can further unlock the model's potential, enabling powerful System-2 thinking. However, there is still a lack of comprehensive surveys for test-time computing scaling. We trace the concept of test-time computing back to System-1 models. In System-1 models, test-time computing addresses distribution shifts and improves robustness and generalization through parameter updating, input modification, representation editing, and output calibration. In System-2 models, it enhances the model's reasoning ability to solve complex problems through repeated sampling, self-correction, and tree search. We organize this survey according to the trend of System-1 to System-2 thinking, highlighting the key role of test-time computing in the transition from System-1 models to weak System-2 models, and then to strong System-2 models. We also point out a few possible future directions.
Authors: Sirine Ayadi, Leon Hetzel, Johanna Sommer, Fabian Theis, Stephan G\"unnemann
Abstract: Effectively designing molecular geometries is essential to advancing pharmaceutical innovations, a domain, which has experienced great attention through the success of generative models and, in particular, diffusion models. However, current molecular diffusion models are tailored towards a specific downstream task and lack adaptability. We introduce UniGuide, a framework for controlled geometric guidance of unconditional diffusion models that allows flexible conditioning during inference without the requirement of extra training or networks. We show how applications such as structure-based, fragment-based, and ligand-based drug design are formulated in the UniGuide framework and demonstrate on-par or superior performance compared to specialised models. Offering a more versatile approach, UniGuide has the potential to streamline the development of molecular generative models, allowing them to be readily used in diverse application scenarios.
Authors: Yuan Cao, Yihan He, Dennis Wu, Hong-Yu Chen, Jianqing Fan, Han Liu
Abstract: Transformers have achieved significant success in various fields, notably excelling in tasks involving sequential data like natural language processing. Despite these achievements, the theoretical understanding of transformers' capabilities remains limited. In this paper, we investigate the theoretical capabilities of transformers to autoregressively generate sequences in Bayesian networks based on in-context maximum likelihood estimation (MLE). Specifically, we consider a setting where a context is formed by a set of independent sequences generated according to a Bayesian network. We demonstrate that there exists a simple transformer model that can (i) estimate the conditional probabilities of the Bayesian network according to the context, and (ii) autoregressively generate a new sample according to the Bayesian network with estimated conditional probabilities. We further demonstrate in extensive experiments that such a transformer does not only exist in theory, but can also be effectively obtained through training. Our analysis highlights the potential of transformers to learn complex probabilistic models and contributes to a better understanding of large language models as a powerful class of sequence generators.
Authors: Zhenglai Li, Jun Wang, Chang Tang, Xinzhong Zhu, Wei Zhang, Xinwang Liu
Abstract: Multi-view clustering (MvC) aims to integrate information from different views to enhance the capability of the model in capturing the underlying data structures. The widely used joint training paradigm in MvC is potentially not fully leverage the multi-view information, since the imbalanced and under-optimized view-specific features caused by the uniform learning objective for all views. For instance, particular views with more discriminative information could dominate the learning process in the joint training paradigm, leading to other views being under-optimized. To alleviate this issue, we first analyze the imbalanced phenomenon in the joint-training paradigm of multi-view clustering from the perspective of gradient descent for each view-specific feature extractor. Then, we propose a novel balanced multi-view clustering (BMvC) method, which introduces a view-specific contrastive regularization (VCR) to modulate the optimization of each view. Concretely, VCR preserves the sample similarities captured from the joint features and view-specific ones into the clustering distributions corresponding to view-specific features to enhance the learning process of view-specific feature extractors. Additionally, a theoretical analysis is provided to illustrate that VCR adaptively modulates the magnitudes of gradients for updating the parameters of view-specific feature extractors to achieve a balanced multi-view learning procedure. In such a manner, BMvC achieves a better trade-off between the exploitation of view-specific patterns and the exploration of view-invariance patterns to fully learn the multi-view information for the clustering task. Finally, a set of experiments are conducted to verify the superiority of the proposed method compared with state-of-the-art approaches both on eight benchmark MvC datasets and two spatially resolved transcriptomics datasets.
Authors: Miguel Carvalho, Bruno Martins
Abstract: Vision-Language Models (VLMs) have recently experienced significant advancements. However, challenges persist in the accurate recognition of fine details within high resolution images, which limits performance in multiple tasks. This work introduces Pheye, a novel architecture that efficiently processes high-resolution images while training fewer parameters than similarly sized VLMs. Notably, Pheye achieves a high efficiency while maintaining strong performance, particularly in tasks that demand fine-grained image understanding and/or the handling of scene-text.
Authors: Jushang Qiu, Lei Wang
Abstract: Skeleton-based action recognition has gained significant attention for its ability to efficiently represent spatiotemporal information in a lightweight format. Most existing approaches use graph-based models to process skeleton sequences, where each pose is represented as a skeletal graph structured around human physical connectivity. Among these, the Spatiotemporal Graph Convolutional Network (ST-GCN) has become a widely used framework. Alternatively, hypergraph-based models, such as the Hyperformer, capture higher-order correlations, offering a more expressive representation of complex joint interactions. A recent advancement, termed Taylor Videos, introduces motion-enhanced skeleton sequences by embedding motion concepts, providing a fresh perspective on interpreting human actions in skeleton-based action recognition. In this paper, we conduct a comprehensive evaluation of both traditional skeleton sequences and Taylor-transformed skeletons using ST-GCN and Hyperformer models on the NTU-60 and NTU-120 datasets. We compare skeletal graph and hypergraph representations, analyzing static poses against motion-injected poses. Our findings highlight the strengths and limitations of Taylor-transformed skeletons, demonstrating their potential to enhance motion dynamics while exposing current challenges in fully using their benefits. This study underscores the need for innovative skeletal modelling techniques to effectively handle motion-rich data and advance the field of action recognition.
Authors: Iustin S\^irbu, Iulia-Renata S\^irbu, Jasmina Bogojeska, Traian Rebedea
Abstract: Medical imaging is crucial for diagnosing, monitoring, and treating medical conditions. The medical reports of radiology images are the primary medium through which medical professionals attest their findings, but their writing is time consuming and requires specialized clinical expertise. The automated generation of radiography reports has thus the potential to improve and standardize patient care and significantly reduce clinicians workload. Through our work, we have designed and evaluated an end-to-end transformer-based method to generate accurate and factually complete radiology reports for X-ray images. Additionally, we are the first to introduce curriculum learning for end-to-end transformers in medical imaging and demonstrate its impact in obtaining improved performance. The experiments have been conducted using the MIMIC-CXR-JPG database, the largest available chest X-ray dataset. The results obtained are comparable with the current state-of-the-art on the natural language generation (NLG) metrics BLEU and ROUGE-L, while setting new state-of-the-art results on F1 examples-averaged, F1-macro and F1-micro metrics for clinical accuracy and on the METEOR metric widely used for NLG.
Authors: Jalisha Jashim Era, Bidyarthi Paul, Tahmid Sattar Aothoi, Mirazur Rahman Zim, Faisal Muhammad Shah
Abstract: Mathematical word problems (MWPs) involve the task of converting textual descriptions into mathematical equations. This poses a significant challenge in natural language processing, particularly for low-resource languages such as Bengali. This paper addresses this challenge by developing an innovative approach to solving Bengali MWPs using transformer-based models, including Basic Transformer, mT5, BanglaT5, and mBART50. To support this effort, the "PatiGonit" dataset was introduced, containing 10,000 Bengali math problems, and these models were fine-tuned to translate the word problems into equations accurately. The evaluation revealed that the mT5 model achieved the highest accuracy of 97.30%, demonstrating the effectiveness of transformer models in this domain. This research marks a significant step forward in Bengali natural language processing, offering valuable methodologies and resources for educational AI tools. By improving math education, it also supports the development of advanced problem-solving skills for Bengali-speaking students.
Authors: Ran Tao, Chong Wang, Hao Chen, Mingjiao Jia, Xiang Shang, Luoyuan Qu, Guoliang Shentu, Yanyu Lu, Yanfeng Huo, Lei Bai, Xianghui Xue, Xiankang Dou
Abstract: Accurate detection of wind fields within the troposphere is essential for atmospheric dynamics research and plays a crucial role in extreme weather forecasting. Coherent Doppler wind lidar (CDWL) is widely regarded as the most suitable technique for high spatial and temporal resolution wind field detection. However, since coherent detection relies heavily on the concentration of aerosol particles, which cause Mie scattering, the received backscattering lidar signal exhibits significantly low intensity at high altitudes. As a result, conventional methods, such as spectral centroid estimation, often fail to produce credible and accurate wind retrieval results in these regions. To address this issue, we propose LWFNet, the first Lidar-based Wind Field (WF) retrieval neural Network, built upon Transformer and the Kolmogorov-Arnold network. Our model is trained solely on targets derived from the traditional wind retrieval algorithm and utilizes radiosonde measurements as the ground truth for test results evaluation. Experimental results demonstrate that LWFNet not only extends the maximum wind field detection range but also produces more accurate results, exhibiting a level of precision that surpasses the labeled targets. This phenomenon, which we refer to as super-accuracy, is explored by investigating the potential underlying factors that contribute to this intriguing occurrence. In addition, we compare the performance of LWFNet with other state-of-the-art (SOTA) models, highlighting its superior effectiveness and capability in high-resolution wind retrieval. LWFNet demonstrates remarkable performance in lidar-based wind field retrieval, setting a benchmark for future research and advancing the development of deep learning models in this domain.
Authors: Simon Park, Abhishek Panigrahi, Yun Cheng, Dingli Yu, Anirudh Goyal, Sanjeev Arora
Abstract: While Vision Language Models (VLMs) are impressive in tasks such as visual question answering (VQA) and image captioning, their ability to apply multi-step reasoning to images has lagged, giving rise to perceptions of modality imbalance or brittleness. Towards systematic study of such issues, we introduce a synthetic framework for assessing the ability of VLMs to perform algorithmic visual reasoning (AVR), comprising three tasks: Table Readout, Grid Navigation, and Visual Analogy. Each has two levels of difficulty, SIMPLE and HARD, and even the SIMPLE versions are difficult for frontier VLMs. We seek strategies for training on the SIMPLE version of the tasks that improve performance on the corresponding HARD task, i.e., S2H generalization. This synthetic framework, where each task also has a text-only version, allows a quantification of the modality imbalance, and how it is impacted by training strategy. Ablations highlight the importance of explicit image-to-text conversion in promoting S2H generalization when using auto-regressive training. We also report results of mechanistic study of this phenomenon, including a measure of gradient alignment that seems to identify training strategies that promote better S2H generalization.
Authors: S. A. Adedayo
Abstract: Characterizing cause-effect relationships in complex systems could be critical to understanding these systems. For many, Granger causality (GC) remains a computational tool of choice to identify causal relations in time series data. Like other causal discovery tools, GC has limitations and has been criticized as a non-causal framework. Here, we addressed one of the recurring criticisms of GC by endowing it with proper causal interpretation. This was achieved by analyzing GC from Reichenbach's Common Cause Principles (RCCPs) and causal Bayesian networks (CBNs) lenses. We showed theoretically and graphically that this reformulation endowed GC with a proper causal interpretation under certain assumptions and achieved satisfactory results on simulation.
Authors: Wen-ran Li, Xavier F. Cadet, David Medina-Ortiz, Mehdi D. Davari, Ramanathan Sowdhamini, Cedric Damour, Yu Li, Alain Miranville, Frederic Cadet
Abstract: Protein design with desirable properties has been a significant challenge for many decades. Generative artificial intelligence is a promising approach and has achieved great success in various protein generation tasks. Notably, diffusion models stand out for their robust mathematical foundations and impressive generative capabilities, offering unique advantages in certain applications such as protein design. In this review, we first give the definition and characteristics of diffusion models and then focus on two strategies: Denoising Diffusion Probabilistic Models and Score-based Generative Models, where DDPM is the discrete form of SGM. Furthermore, we discuss their applications in protein design, peptide generation, drug discovery, and protein-ligand interaction. Finally, we outline the future perspectives of diffusion models to advance autonomous protein design and engineering. The E(3) group consists of all rotations, reflections, and translations in three-dimensions. The equivariance on the E(3) group can keep the physical stability of the frame of each amino acid as much as possible, and we reflect on how to keep the diffusion model E(3) equivariant for protein generation.
Authors: Nayeli A. Rodr\'iguez-Briones, Daniel K. Park
Abstract: This work introduces an approach rooted in quantum thermodynamics to enhance sampling efficiency in quantum machine learning (QML). We propose conceptualizing quantum supervised learning as a thermodynamic cooling process. Building on this concept, we develop a quantum refrigerator protocol that enhances sample efficiency during training and prediction without the need for Grover iterations or quantum phase estimation. Inspired by heat-bath algorithmic cooling protocols, our method alternates entropy compression and thermalization steps to decrease the entropy of qubits, increasing polarization towards the dominant bias. This technique minimizes the computational overhead associated with estimating classification scores and gradients, presenting a practical and efficient solution for QML algorithms compatible with noisy intermediate-scale quantum devices.
Authors: Binita Saha, Utsha Saha, Muhammad Zubair Malik
Abstract: This work presents a novel architecture for building Retrieval-Augmented Generation (RAG) systems to improve Question Answering (QA) tasks from a target corpus. Large Language Models (LLMs) have revolutionized the analyzing and generation of human-like text. These models rely on pre-trained data and lack real-time updates unless integrated with live data tools. RAG enhances LLMs by integrating online resources and databases to generate contextually appropriate responses. However, traditional RAG still encounters challenges like information dilution and hallucinations when handling vast amounts of data. Our approach addresses these challenges by converting corpora into a domain-specific dataset and RAG architecture is constructed to generate responses from the target document. We introduce QuIM-RAG (Question-to-question Inverted Index Matching), a novel approach for the retrieval mechanism in our system. This strategy generates potential questions from document chunks and matches these with user queries to identify the most relevant text chunks for generating accurate answers. We have implemented our RAG system on top of the open-source Meta-LLaMA3-8B-instruct model by Meta Inc. that is available on Hugging Face. We constructed a custom corpus of 500+ pages from a high-traffic website accessed thousands of times daily for answering complex questions, along with manually prepared ground truth QA for evaluation. We compared our approach with traditional RAG models using BERT-Score and RAGAS, state-of-the-art metrics for evaluating LLM applications. Our evaluation demonstrates that our approach outperforms traditional RAG architectures on both metrics.
Authors: Mehran Shoushtari Moghadam, Sercan Aygun, M. Hassan Najafi
Abstract: Data encoding is a fundamental step in emerging computing paradigms, particularly in stochastic computing (SC) and hyperdimensional computing (HDC), where it plays a crucial role in determining the overall system performance and hardware cost efficiency. This study presents an advanced encoding strategy that leverages a hardware-friendly class of low-discrepancy (LD) sequences, specifically powers-of-2 bases of Van der Corput (VDC) sequences (VDC-2^n), as sources for random number generation. Our approach significantly enhances the accuracy and efficiency of SC and HDC systems by addressing challenges associated with randomness. By employing LD sequences, we improve correlation properties and reduce hardware complexity. Experimental results demonstrate significant improvements in accuracy and energy savings for SC and HDC systems. Our solution provides a robust framework for integrating SC and HDC in resource-constrained environments, paving the way for efficient and scalable AI implementations.
Authors: Ji Cao, Tongya Zheng, Qinghong Guo, Yu Wang, Junshu Dai, Shunyu Liu, Jie Yang, Jie Song, Mingli Song
Abstract: Trajectory generation has garnered significant attention from researchers in the field of spatio-temporal analysis, as it can generate substantial synthesized human mobility trajectories that enhance user privacy and alleviate data scarcity. However, existing trajectory generation methods often focus on improving trajectory generation quality from a singular perspective, lacking a comprehensive semantic understanding across various scales. Consequently, we are inspired to develop a HOlistic SEmantic Representation (HOSER) framework for navigational trajectory generation. Given an origin-and-destination (OD) pair and the starting time point of a latent trajectory, we first propose a Road Network Encoder to expand the receptive field of road- and zone-level semantics. Second, we design a Multi-Granularity Trajectory Encoder to integrate the spatio-temporal semantics of the generated trajectory at both the point and trajectory levels. Finally, we employ a Destination-Oriented Navigator to seamlessly integrate destination-oriented guidance. Extensive experiments on three real-world datasets demonstrate that HOSER outperforms state-of-the-art baselines by a significant margin. Moreover, the model's performance in few-shot learning and zero-shot learning scenarios further verifies the effectiveness of our holistic semantic representation.
Authors: Kyungmin Kim, SangHun Im, GiBaeg Kim, Heung-Seon Oh
Abstract: Text augmentation (TA) is a critical technique for text classification, especially in few-shot settings. This paper introduces a novel LLM-based TA method, TARDiS, to address challenges inherent in the generation and alignment stages of two-stage TA methods. For the generation stage, we propose two generation processes, SEG and CEG, incorporating multiple class-specific prompts to enhance diversity and separability. For the alignment stage, we introduce a class adaptation (CA) method to ensure that generated examples align with their target classes through verification and modification. Experimental results demonstrate TARDiS's effectiveness, outperforming state-of-the-art LLM-based TA methods in various few-shot text classification tasks. An in-depth analysis confirms the detailed behaviors at each stage.
Authors: Shengming Zhang, Le Zhang, Jingbo Zhou, Hui Xiong
Abstract: Link prediction in heterogeneous networks is crucial for understanding the intricacies of network structures and forecasting their future developments. Traditional methodologies often face significant obstacles, including over-smoothing-wherein the excessive aggregation of node features leads to the loss of critical structural details-and a dependency on human-defined meta-paths, which necessitate extensive domain knowledge and can be inherently restrictive. These limitations hinder the effective prediction and analysis of complex heterogeneous networks. In response to these challenges, we propose the Contrastive Heterogeneous grAph Transformer (CHAT). CHAT introduces a novel sampling-based graph transformer technique that selectively retains nodes of interest, thereby obviating the need for predefined meta-paths. The method employs an innovative connection-aware transformer to encode node sequences and their interconnections with high fidelity, guided by a dual-faceted loss function specifically designed for heterogeneous network link prediction. Additionally, CHAT incorporates an ensemble link predictor that synthesizes multiple samplings to achieve enhanced prediction accuracy. We conducted comprehensive evaluations of CHAT using three distinct drug-target interaction (DTI) datasets. The empirical results underscore CHAT's superior performance, outperforming both general-task approaches and models specialized in DTI prediction. These findings substantiate the efficacy of CHAT in addressing the complex problem of link prediction in heterogeneous networks.
Authors: Wenzhi Gao, Dongdong Ge, Chenyu Xue, Chunlin Sun, Yinyu Ye
Abstract: Online linear programming plays an important role in both revenue management and resource allocation, and recent research has focused on developing efficient first-order online learning algorithms. Despite the empirical success of first-order methods, they typically achieve a regret no better than $\mathcal{O} ( \sqrt{T} )$, which is suboptimal compared to the $\mathcal{O} (\log T)$ bound guaranteed by the state-of-the-art linear programming (LP)-based online algorithms. This paper establishes a general framework that improves upon the $\mathcal{O} ( \sqrt{T} )$ result when the LP dual problem exhibits certain error bound conditions. For the first time, we show that first-order learning algorithms achieve $o( \sqrt{T} )$ regret in the continuous support setting and $\mathcal{O} (\log T)$ regret in the finite support setting beyond the non-degeneracy assumption. Our results significantly improve the state-of-the-art regret results and provide new insights for sequential decision-making.
Authors: Sugandha Saxena, S. N. Prasad, Ashwin M Polnaya, Shweta Agarwala
Abstract: Advances in healthcare research have significantly enhanced our understanding of disease mechanisms, diagnostic precision, and therapeutic options. Yet, lung cancer remains one of the leading causes of cancer-related mortality worldwide due to challenges in early and accurate diagnosis. While current lung cancer detection models show promise, there is considerable potential for further improving the accuracy for timely intervention. To address this challenge, we introduce a hybrid deep convolution model leveraging transfer learning, named the Maximum Sensitivity Neural Network (MSNN). MSNN is designed to improve the precision of lung cancer detection by refining sensitivity and specificity. This model has surpassed existing deep learning approaches through experimental validation, achieving an accuracy of 98% and a sensitivity of 97%. By overlaying sensitivity maps onto lung Computed Tomography (CT) scans, it enables the visualization of regions most indicative of malignant or benign classifications. This innovative method demonstrates exceptional performance in distinguishing lung cancer with minimal false positives, thereby enhancing the accuracy of medical diagnoses.
Authors: Niloufar Eghbali, Hassan Bagher-Ebadian, Tuka Alhanai, Mohammad M. Ghassemi
Abstract: Vision Transformers (ViTs) have shown promise in medical image semantic segmentation (MISS) by capturing long-range correlations. However, ViTs often struggle to model local spatial information effectively, which is essential for accurately segmenting fine anatomical details, particularly when applied to small datasets without extensive pre-training. We introduce Gabor and Laplacian of Gaussian Convolutional Swin Network (GLoG-CSUnet), a novel architecture enhancing Transformer-based models by incorporating learnable radiomic features. This approach integrates dynamically adaptive Gabor and Laplacian of Gaussian (LoG) filters to capture texture, edge, and boundary information, enhancing the feature representation processed by the Transformer model. Our method uniquely combines the long-range dependency modeling of Transformers with the texture analysis capabilities of Gabor and LoG features. Evaluated on the Synapse multi-organ and ACDC cardiac segmentation datasets, GLoG-CSUnet demonstrates significant improvements over state-of-the-art models, achieving a 1.14\% increase in Dice score for Synapse and 0.99\% for ACDC, with minimal computational overhead (only 15 and 30 additional parameters, respectively). GLoG-CSUnet's flexible design allows integration with various base models, offering a promising approach for incorporating radiomics-inspired feature extraction in Transformer architectures for medical image analysis. The code implementation is available on GitHub at: https://github.com/HAAIL/GLoG-CSUnet.
Authors: Ye Lin, Jiwei Jia, Young Ju Lee, Ran Zhang
Abstract: Greedy algorithms, particularly the orthogonal greedy algorithm (OGA), have proven effective in training shallow neural networks for fitting functions and solving partial differential equations (PDEs). In this paper, we extend the application of OGA to the tasks of linear operator learning, which is equivalent to learning the kernel function through integral transforms. Firstly, a novel greedy algorithm is developed for kernel estimation rate in a new semi-inner product, which can be utilized to approximate the Green's function of linear PDEs from data. Secondly, we introduce the OGA for point-wise kernel estimation to further improve the approximation rate, achieving orders of accuracy improvement across various tasks and baseline models. In addition, we provide a theoretical analysis on the kernel estimation problem and the optimal approximation rates for both algorithms, establishing their efficacy and potential for future applications in PDEs and operator learning tasks.
Authors: Kunwoong Kim, Insung Kong, Jongjin Lee, Minwoo Chae, Sangchul Park, Yongdai Kim
Abstract: Group fairness requires that different protected groups, characterized by a given sensitive attribute, receive equal outcomes overall. Typically, the level of group fairness is measured by the statistical gap between predictions from different protected groups. In this study, we reveal an implicit property of existing group fairness measures, which provides an insight into how the group-fair models behave. Then, we develop a new group-fair constraint based on this implicit property to learn group-fair models. To do so, we first introduce a notable theoretical observation: every group-fair model has an implicitly corresponding transport map between the input spaces of each protected group. Based on this observation, we introduce a new group fairness measure termed Matched Demographic Parity (MDP), which quantifies the averaged gap between predictions of two individuals (from different protected groups) matched by a given transport map. Then, we prove that any transport map can be used in MDP to learn group-fair models, and develop a novel algorithm called Fairness Through Matching (FTM), which learns a group-fair model using MDP constraint with an user-specified transport map. We specifically propose two favorable types of transport maps for MDP, based on the optimal transport theory, and discuss their advantages. Experiments reveal that FTM successfully trains group-fair models with certain desirable properties by choosing the transport map accordingly.
Authors: Saba Fathi Rabooki, Bowen Li, Falih Gozi Febrinanto, Ciyuan Peng, Elham Naghizade, Fengling Han, Feng Xia
Abstract: Cyber-physical-social systems (CPSSs) have emerged in many applications over recent decades, requiring increased attention to security concerns. The rise of sophisticated threats like Advanced Persistent Threats (APTs) makes ensuring security in CPSSs particularly challenging. Provenance graph analysis has proven effective for tracing and detecting anomalies within systems, but the sheer size and complexity of these graphs hinder the efficiency of existing methods, especially those relying on graph neural networks (GNNs). To address these challenges, we present GraphDART, a modular framework designed to distill provenance graphs into compact yet informative representations, enabling scalable and effective anomaly detection. GraphDART can take advantage of diverse graph distillation techniques, including classic and modern graph distillation methods, to condense large provenance graphs while preserving essential structural and contextual information. This approach significantly reduces computational overhead, allowing GNNs to learn from distilled graphs efficiently and enhance detection performance. Extensive evaluations on benchmark datasets demonstrate the robustness of GraphDART in detecting malicious activities across cyber-physical-social systems. By optimizing computational efficiency, GraphDART provides a scalable and practical solution to safeguard interconnected environments against APTs.
Authors: Yee Chun Tsoi, Yu Ting Kwok, Ming Chun Lam, Wai Kin Wong
Abstract: In the Hong Kong Observatory, the Analogue Forecast System (AFS) for precipitation has been providing useful reference in predicting possible daily rainfall scenarios for the next 9 days, by identifying historical cases with similar weather patterns to the latest output from the deterministic model of the European Centre for Medium-Range Weather Forecasts (ECMWF). Recent advances in machine learning allow more sophisticated models to be trained using historical data and the patterns of high-impact weather events to be represented more effectively. As such, an enhanced AFS has been developed using the deep learning technique autoencoder. The datasets of the fifth generation of the ECMWF Reanalysis (ERA5) are utilised where more meteorological elements in higher horizontal, vertical and temporal resolutions are available as compared to the previous ECMWF reanalysis products used in the existing AFS. The enhanced AFS features four major steps in generating the daily rain class forecasts: (1) preprocessing of gridded ERA5 and ECMWF model forecast, (2) feature extraction by the pretrained autoencoder, (3) application of optimised feature weightings based on historical cases, and (4) calculation of the final rain class from a weighted ensemble of top analogues. The enhanced AFS demonstrates a consistent and superior performance over the existing AFS, especially in capturing heavy rain cases, during the verification period from 2019 to 2022. This paper presents the detailed formulation of the enhanced AFS and discusses its advantages and limitations in supporting precipitation forecasting in Hong Kong.
Authors: Jian Jiang, Long Chen, Yueying Zhu, Yazhou Shi, Huahai Qiu, Bengong Zhang, Tianshou Zhou, Guo-Wei Wei
Abstract: Anesthetics are crucial in surgical procedures and therapeutic interventions, but they come with side effects and varying levels of effectiveness, calling for novel anesthetic agents that offer more precise and controllable effects. Targeting Gamma-aminobutyric acid (GABA) receptors, the primary inhibitory receptors in the central nervous system, could enhance their inhibitory action, potentially reducing side effects while improving the potency of anesthetics. In this study, we introduce a proteomic learning of GABA receptor-mediated anesthesia based on 24 GABA receptor subtypes by considering over 4000 proteins in protein-protein interaction (PPI) networks and over 1.5 millions known binding compounds. We develop a corresponding drug-target interaction network to identify potential lead compounds for novel anesthetic design. To ensure robust proteomic learning predictions, we curated a dataset comprising 136 targets from a pool of 980 targets within the PPI networks. We employed three machine learning algorithms, integrating advanced natural language processing (NLP) models such as pretrained transformer and autoencoder embeddings. Through a comprehensive screening process, we evaluated the side effects and repurposing potential of over 180,000 drug candidates targeting the GABRA5 receptor. Additionally, we assessed the ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties of these candidates to identify those with near-optimal characteristics. This approach also involved optimizing the structures of existing anesthetics. Our work presents an innovative strategy for the development of new anesthetic drugs, optimization of anesthetic use, and deeper understanding of potential anesthesia-related side effects.
Authors: Kuldeep Kurte, Kedar Kulkarni
Abstract: In this paper, we present an enhanced Convolutional Neural Network (CNN)-based rooftop solar photovoltaic (PV) panel detection approach using satellite images. We propose to use pre-trained CNN-based model to extract the local convolutional features of rooftops. These local features are then combined using the Vectors of Locally Aggregated Descriptors (VLAD) technique to obtain rooftop-level global features, which are then used to train traditional Machine Learning (ML) models to identify rooftop images that do and do not contain PV panels. On the dataset used in this study, the proposed approach achieved rooftop-PV classification scores exceeding the predefined threshold of 0.9 across all three cities for each of the feature extractor networks evaluated. Moreover, we propose a 3-phase approach to enable efficient utilization of the previously trained models on a new city or region with limited labelled data. We illustrate the effectiveness of this 3-phase approach for multi-city rooftop-PV detection task.
Authors: Qingyao Ai, Jingtao Zhan, Yiqun Liu
Abstract: The chapter discusses the foundational impact of modern generative AI models on information access (IA) systems. In contrast to traditional AI, the large-scale training and superior data modeling of generative AI models enable them to produce high-quality, human-like responses, which brings brand new opportunities for the development of IA paradigms. In this chapter, we identify and introduce two of them in details, i.e., information generation and information synthesis. Information generation allows AI to create tailored content addressing user needs directly, enhancing user experience with immediate, relevant outputs. Information synthesis leverages the ability of generative AI to integrate and reorganize existing information, providing grounded responses and mitigating issues like model hallucination, which is particularly valuable in scenarios requiring precision and external knowledge. This chapter delves into the foundational aspects of generative models, including architecture, scaling, and training, and discusses their applications in multi-modal scenarios. Additionally, it examines the retrieval-augmented generation paradigm and other methods for corpus modeling and understanding, demonstrating how generative AI can enhance information access systems. It also summarizes potential challenges and fruitful directions for future studies.
Authors: Xia Chen, Yugen Du, Guoxing Tang, Yingwei Luo, Benchi Ma
Abstract: As the homogenization of Web services becomes more and more common, the difficulty of service recommendation is gradually increasing. How to predict Quality of Service (QoS) more efficiently and accurately becomes an important challenge for service recommendation. Considering the excellent role of reputation and deep learning (DL) techniques in the field of QoS prediction, we propose a reputation and DL based QoS prediction network, RAHN, which contains the Reputation Calculation Module (RCM), the Latent Feature Extraction Module (LFEM), and the QoS Prediction Hourglass Network (QPHN). RCM obtains the user reputation and the service reputation by using a clustering algorithm and a Logit model. LFEM extracts latent features from known information to form an initial latent feature vector. QPHN aggregates latent feature vectors with different scales by using Attention Mechanism, and can be stacked multiple times to obtain the final latent feature vector for prediction. We evaluate RAHN on a real QoS dataset. The experimental results show that the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) of RAHN are smaller than the six baseline methods.
Authors: Yubo Wang, Haoyang Li, Fei Teng, Lei Chen
Abstract: Text classification is a fundamental task in natural language processing, pivotal to various applications such as query optimization, data integration, and schema matching. While neural network-based models, such as CNN and BERT, have demonstrated remarkable performance in text classification, their effectiveness heavily relies on abundant labeled training data. This dependency makes these models less effective in dynamic few-shot text classification, where labeled data is scarce, and target labels frequently evolve based on application needs. Recently, large language models (LLMs) have shown promise due to their extensive pretraining and contextual understanding. Current approaches provide LLMs with text inputs, candidate labels, and additional side information (e.g., descriptions) to predict text labels. However, their effectiveness is hindered by the increased input size and the noise introduced through side information processing. To address these limitations, we propose a graph-based online retrieval-augmented generation framework, namely GORAG, for dynamic few-shot text classification. GORAG constructs and maintains an adaptive information graph by extracting side information across all target texts, rather than treating each input independently. It employs a weighted edge mechanism to prioritize the importance and reliability of extracted information and dynamically retrieves relevant context using a minimum-cost spanning tree tailored for each text input. Empirical evaluations demonstrate that GORAG outperforms existing approaches by providing more comprehensive and accurate contextual information.
Authors: Yuxin Ma, Zherui Zhang, Ran Cheng, Yaochu Jin, Kay Chen Tan
Abstract: In the domain of multi-objective optimization, evolutionary algorithms are distinguished by their capability to generate a diverse population of solutions that navigate the trade-offs inherent among competing objectives. This has catalyzed the ascension of evolutionary multi-objective optimization (EMO) as a prevalent approach. Despite the effectiveness of the EMO paradigm, the analysis of resultant solution sets presents considerable challenges. This is primarily attributed to the high-dimensional nature of the data and the constraints imposed by static visualization methods, which frequently culminate in visual clutter and impede interactive exploratory analysis. To address these challenges, this paper introduces ParetoLens, a visual analytics framework specifically tailored to enhance the inspection and exploration of solution sets derived from the multi-objective evolutionary algorithms. Utilizing a modularized, algorithm-agnostic design, ParetoLens enables a detailed inspection of solution distributions in both decision and objective spaces through a suite of interactive visual representations. This approach not only mitigates the issues associated with static visualizations but also supports a more nuanced and flexible analysis process. The usability of the framework is evaluated through case studies and expert interviews, demonstrating its potential to uncover complex patterns and facilitate a deeper understanding of multi-objective optimization solution sets. A demo website of ParetoLens is available at https://dva-lab.org/paretolens/.
Authors: Hui Sun, Shiyin Lu, Huanyu Wang, Qing-Guo Chen, Zhao Xu, Weihua Luo, Kaifu Zhang, Ming Li
Abstract: Video large language models (Video-LLMs) have made significant progress in understanding videos. However, processing multiple frames leads to lengthy visual token sequences, presenting challenges such as the limited context length cannot accommodate the entire video, and the inclusion of irrelevant frames hinders visual perception. Hence, effective frame selection is crucial. This paper emphasizes that frame selection should follow three key principles: query relevance, list-wise diversity, and sequentiality. Existing methods, such as uniform frame sampling and query-frame matching, do not capture all of these principles. Thus, we propose Markov decision determinantal point process with dynamic programming (MDP3) for frame selection, a training-free and model-agnostic method that can be seamlessly integrated into existing Video-LLMs. Our method first estimates frame similarities conditioned on the query using a conditional Gaussian kernel within the reproducing kernel Hilbert space~(RKHS). We then apply the determinantal point process~(DPP) to the similarity matrix to capture both query relevance and list-wise diversity. To incorporate sequentiality, we segment the video and apply DPP within each segment, conditioned on the preceding segment selection, modeled as a Markov decision process~(MDP) for allocating selection sizes across segments. Theoretically, MDP3 provides a \((1 - 1/e)\)-approximate solution to the NP-hard list-wise frame selection problem with pseudo-polynomial time complexity, demonstrating its efficiency. Empirically, MDP3 significantly outperforms existing methods, verifying its effectiveness and robustness.
Authors: Sahar Salimpour, Jorge Pe\~na-Queralta, Diego Paez-Granados, Jukka Heikkonen, Tomi Westerlund
Abstract: Unprecedented agility and dexterous manipulation have been demonstrated with controllers based on deep reinforcement learning (RL), with a significant impact on legged and humanoid robots. Modern tooling and simulation platforms, such as NVIDIA Isaac Sim, have been enabling such advances. This article focuses on demonstrating the applications of Isaac in local planning and obstacle avoidance as one of the most fundamental ways in which a mobile robot interacts with its environments. Although there is extensive research on proprioception-based RL policies, the article highlights less standardized and reproducible approaches to exteroception. At the same time, the article aims to provide a base framework for end-to-end local navigation policies and how a custom robot can be trained in such simulation environment. We benchmark end-to-end policies with the state-of-the-art Nav2, navigation stack in Robot Operating System (ROS). We also cover the sim-to-real transfer process by demonstrating zero-shot transferability of policies trained in the Isaac simulator to real-world robots. This is further evidenced by the tests with different simulated robots, which show the generalization of the learned policy. Finally, the benchmarks demonstrate comparable performance to Nav2, opening the door to quick deployment of state-of-the-art end-to-end local planners for custom robot platforms, but importantly furthering the possibilities by expanding the state and action spaces or task definitions for more complex missions. Overall, with this article we introduce the most important steps, and aspects to consider, in deploying RL policies for local path planning and obstacle avoidance with Isaac Sim training, Gazebo testing, and ROS 2 for real-time inference in real robots. The code is available at https://github.com/sahars93/RL-Navigation.
Authors: Jonathan Courtois, Beno\^it Miramond, Alain Pegatoquet
Abstract: With the growing interest in on On-orbit servicing (OOS) and Active Debris Removal (ADR) missions, spacecraft poses estimation algorithms are being developed using deep learning to improve the precision of this complex task and find the most efficient solution. With the advances of bio-inspired low-power solutions, such a spiking neural networks and event-based processing and cameras, and their recent work for space applications, we propose to investigate the feasibility of a fully event-based solution to improve event-based pose estimation for spacecraft. In this paper, we address the first event-based dataset SEENIC with real event frames captured by an event-based camera on a testbed. We show the methods and results of the first event-based solution for this use case, where our small spiking end-to-end network (S2E2) solution achieves interesting results over 21cm position error and 14degree rotation error, which is the first step towards fully event-based processing for embedded spacecraft pose estimation.
Authors: Andrew Ma, Owen Dugan, Marin Solja\v{c}i\'c
Abstract: In solid-state materials science, substantial efforts have been devoted to the calculation and modeling of the electronic band gap. While a wide range of ab initio methods and machine learning algorithms have been created that can predict this quantity, the development of new computational approaches for studying the band gap remains an active area of research. Here we introduce a simple machine learning model for predicting the band gap using only the chemical composition of the crystalline material. To motivate the form of the model, we first analyze the empirical distribution of the band gap, which sheds new light on its atypical statistics. Specifically, our analysis enables us to frame band gap prediction as a task of modeling a mixed random variable, and we design our model accordingly. Our model formulation incorporates thematic ideas from chemical heuristic models for other material properties in a manner that is suited towards the band gap modeling task. The model has exactly one parameter corresponding to each element, which is fit using data. To predict the band gap for a given material, the model computes a weighted average of the parameters associated with its constituent elements and then takes the maximum of this quantity and zero. The model provides heuristic chemical interpretability by intuitively capturing the associations between the band gap and individual chemical elements.
Authors: Debangshu Chowdhury, Souvik Chakraborty
Abstract: Time-delayed differential equations (TDDEs) are widely used to model complex dynamic systems where future states depend on past states with a delay. However, inferring the underlying TDDEs from observed data remains a challenging problem due to the inherent nonlinearity, uncertainty, and noise in real-world systems. Conventional equation discovery methods often exhibit limitations when dealing with large time delays, relying on deterministic techniques or optimization-based approaches that may struggle with scalability and robustness. In this paper, we present BayTiDe - Bayesian Approach for Discovering Time-Delayed Differential Equations from Data, that is capable of identifying arbitrarily large values of time delay to an accuracy that is directly proportional to the resolution of the data input to it. BayTiDe leverages Bayesian inference combined with a sparsity-promoting discontinuous spike-and-slab prior to accurately identify time-delayed differential equations. The approach accommodates arbitrarily large time delays with accuracy proportional to the input data resolution, while efficiently narrowing the search space to achieve significant computational savings. We demonstrate the efficiency and robustness of BayTiDe through a range of numerical examples, validating its ability to recover delayed differential equations from noisy data.
Authors: Ryan Cory-Wright, Jean Pauphilet
Abstract: Inspired by the impact of the Goemans-Williamson algorithm on combinatorial optimization, we construct an analogous relax-then-sample strategy for low-rank optimization problems. First, for orthogonally constrained quadratic optimization problems, we derive a semidefinite relaxation and a randomized rounding scheme, which obtains provably near-optimal solutions, mimicking the blueprint from Goemans and Williamson for the Max-Cut problem. We then extend our approach to generic low-rank optimization problems by developing new semidefinite relaxations that are both tighter and more broadly applicable than those in prior works. Although our original proposal introduces large semidefinite matrices as decision variables, we show that most of the blocks in these matrices can be safely omitted without altering the optimal value, hence improving the scalability of our approach. Using several examples (including matrix completion, basis pursuit, and reduced-rank regression), we show how to reduce the size of our relaxation even further. Finally, we numerically illustrate the effectiveness and scalability of our relaxation and our sampling scheme on orthogonally constrained quadratic optimization and matrix completion problems.
Authors: Samuel J. Gershman, Ila Fiete, Kazuki Irie
Abstract: Classical models of memory in psychology and neuroscience rely on similarity-based retrieval of stored patterns, where similarity is a function of retrieval cues and the stored patterns. While parsimonious, these models do not allow distinct representations for storage and retrieval, despite their distinct computational demands. Key-value memory systems, in contrast, distinguish representations used for storage (values) and those used for retrieval (keys). This allows key-value memory systems to optimize simultaneously for fidelity in storage and discriminability in retrieval. We review the computational foundations of key-value memory, its role in modern machine learning systems, related ideas from psychology and neuroscience, applications to a number of empirical puzzles, and possible biological implementations.
Authors: Alexander Merkov, David Rohde
Abstract: This paper provides a formalism for an important class of causal inference problems inspired by user-advertiser interaction in online advertiser. Then this formalism is specialized to an extension of temporal marked point processes and the neural point processes are suggested as practical solutions to some interesting special cases.
Authors: Timothy Schauml\"offel, Arthur Aubret, Gemma Roig, Jochen Triesch
Abstract: Recent self-supervised learning (SSL) models trained on human-like egocentric visual inputs substantially underperform on image recognition tasks compared to humans. These models train on raw, uniform visual inputs collected from head-mounted cameras. This is different from humans, as the anatomical structure of the retina and visual cortex relatively amplifies the central visual information, i.e. around humans' gaze location. This selective amplification in humans likely aids in forming object-centered visual representations. Here, we investigate whether focusing on central visual information boosts egocentric visual object learning. We simulate 5-months of egocentric visual experience using the large-scale Ego4D dataset and generate gaze locations with a human gaze prediction model. To account for the importance of central vision in humans, we crop the visual area around the gaze location. Finally, we train a time-based SSL model on these modified inputs. Our experiments demonstrate that focusing on central vision leads to better object-centered representations. Our analysis shows that the SSL model leverages the temporal dynamics of the gaze movements to build stronger visual representations. Overall, our work marks a significant step toward bio-inspired learning of visual representations.
Authors: Elouan Argouarc'h, Fran\c{c}ois Desbouvries, Eric Barat, Eiji Kawasaki, Thomas Dautremer
Abstract: This paper proposes an extension of standard mixture stochastic models, by replacing the constant mixture weights with functional weights defined using a classifier. Classifier Weighted Mixtures enable straightforward density evaluation, explicit sampling, and enhanced expressivity in variational estimation problems, without increasing the number of components nor the complexity of the mixture components.
Authors: William Consagra, Zhiling Gu, Zhengwu Zhang
Abstract: We propose a novel deep neural network methodology for density estimation on product Riemannian manifold domains. In our approach, the network directly parameterizes the unknown density function and is trained using a penalized maximum likelihood framework, with a penalty term formed using manifold differential operators. The network architecture and estimation algorithm are carefully designed to handle the challenges of high-dimensional product manifold domains, effectively mitigating the curse of dimensionality that limits traditional kernel and basis expansion estimators, as well as overcoming the convergence issues encountered by non-specialized neural network methods. Extensive simulations and a real-world application to brain structural connectivity data highlight the clear advantages of our method over the competing alternatives.
Authors: Arda Atal{\i}k, Sumit Chopra, Daniel K. Sodickson
Abstract: Reducing MRI scan times can improve patient care and lower healthcare costs. Many acceleration methods are designed to reconstruct diagnostic-quality images from limited sets of acquired $\textit{k}$-space data. This task can be framed as a linear inverse problem (LIP), where, as a result of undersampling, the forward operator may become rank-deficient or exhibit small singular values. This results in ambiguities in reconstruction, in which multiple generally incorrect or non-diagnostic images can map to the same acquired data. To address such ambiguities, it is crucial to incorporate prior knowledge, for example in the form of regularization. Another form of prior knowledge less commonly used in medical imaging is contextual side information garnered from other sources than the current acquisition. Here, we propose the $\textbf{T}$rust-$\textbf{G}$uided $\textbf{V}$ariational $\textbf{N}$etwork $\textbf{(TGVN)}$, a novel end-to-end deep learning framework that effectively integrates side information into LIPs. TGVN eliminates undesirable solutions from the ambiguous space of the forward operator while remaining faithful to the acquired data. We demonstrate its effectiveness in multi-coil, multi-contrast MR image reconstruction, where incomplete or low-quality measurements from one contrast are used as side information to reconstruct high-quality images of another contrast from heavily under-sampled data. Our method is robust across different contrasts, anatomies, and field strengths. Compared to baselines that also utilize side information, TGVN achieves superior image quality at challenging under-sampling levels, drastically speeding up acquisition while minimizing hallucinations. Our approach is also versatile enough to incorporate many different types of side information (including previous scans or even text) into any LIP.
Authors: Sheldon Z. Soudin
Abstract: Making sense of theory choice in normal and across extraordinary science is central to philosophy of science. The emergence of machine learning models has the potential to act as a wrench in the gears of current debates. In this paper, I will attempt to reconstruct the main movements that lead to and came out of Putnam's critical and explanatory tendency distinction, argue for the biconditional necessity of the tendencies, and conceptualize that wrench through a machine learning interpretation of my claim.
Authors: Dichucheng Li, Yongyi Zang, Qiuqiang Kong
Abstract: Automatic Music Transcription (AMT), aiming to get musical notes from raw audio, typically uses frame-level systems with piano-roll outputs or language model (LM)-based systems with note-level predictions. However, frame-level systems require manual thresholding, while the LM-based systems struggle with long sequences. In this paper, we propose a hybrid method combining pre-trained roll-based encoders with an LM decoder to leverage the strengths of both methods. Besides, our approach employs a hierarchical prediction strategy, first predicting onset and pitch, then velocity, and finally offset. The hierarchical prediction strategy reduces computational costs by breaking down long sequences into different hierarchies. Evaluated on two benchmark roll-based encoders, our method outperforms traditional piano-roll outputs 0.01 and 0.022 in onset-offset-velocity F1 score, demonstrating its potential as a performance-enhancing plug-in for arbitrary roll-based music transcription encoder. We release the code of this work at https://github.com/yongyizang/AMT_train.
Authors: Jingyi Wang, Ying Chen, Paolo Giudici
Abstract: We propose Group Shapley, a metric that extends the classical individual-level Shapley value framework to evaluate the importance of feature groups, addressing the structured nature of predictors commonly found in business and economic data. More importantly, we develop a significance testing procedure based on a three-cumulant chi-square approximation and establish the asymptotic properties of the test statistics for Group Shapley values. Our approach can effectively handle challenging scenarios, including sparse or skewed distributions and small sample sizes, outperforming alternative tests such as the Wald test. Simulations confirm that the proposed test maintains robust empirical size and demonstrates enhanced power under diverse conditions. To illustrate the method's practical relevance in advancing Explainable AI, we apply our framework to bond recovery rate predictions using a global dataset (1996-2023) comprising 2,094 observations and 98 features, grouped into 16 subgroups and five broader categories: bond characteristics, firm fundamentals, industry-specific factors, market-related variables, and macroeconomic indicators. Our results identify the market-related variables group as the most influential. Furthermore, Lorenz curves and Gini indices reveal that Group Shapley assigns feature importance more equitably compared to individual Shapley values.
Authors: Guy Yariv, Yuval Kirstain, Amit Zohar, Shelly Sheynin, Yaniv Taigman, Yossi Adi, Sagie Benaim, Adam Polyak
Abstract: We consider the task of Image-to-Video (I2V) generation, which involves transforming static images into realistic video sequences based on a textual description. While recent advancements produce photorealistic outputs, they frequently struggle to create videos with accurate and consistent object motion, especially in multi-object scenarios. To address these limitations, we propose a two-stage compositional framework that decomposes I2V generation into: (i) An explicit intermediate representation generation stage, followed by (ii) A video generation stage that is conditioned on this representation. Our key innovation is the introduction of a mask-based motion trajectory as an intermediate representation, that captures both semantic object information and motion, enabling an expressive but compact representation of motion and semantics. To incorporate the learned representation in the second stage, we utilize object-level attention objectives. Specifically, we consider a spatial, per-object, masked-cross attention objective, integrating object-specific prompts into corresponding latent space regions and a masked spatio-temporal self-attention objective, ensuring frame-to-frame consistency for each object. We evaluate our method on challenging benchmarks with multi-object and high-motion scenarios and empirically demonstrate that the proposed method achieves state-of-the-art results in temporal coherence, motion realism, and text-prompt faithfulness. Additionally, we introduce \benchmark, a new challenging benchmark for single-object and multi-object I2V generation, and demonstrate our method's superiority on this benchmark. Project page is available at https://guyyariv.github.io/TTM/.
Authors: Philipp Teutsch, Philipp Pfeffer, Mohammad Sharifi Ghazijahani, Christian Cierpka, J\"org Schumacher, Patrick M\"ader
Abstract: In recent years, data-driven deep learning models have gained significant interest in the analysis of turbulent dynamical systems. Within the context of reduced-order models (ROMs), convolutional autoencoders (CAEs) pose a universally applicable alternative to conventional approaches. They can learn nonlinear transformations directly from data, without prior knowledge of the system. However, the features generated by such models lack interpretability. Thus, the resulting model is a black-box which effectively reduces the complexity of the system, but does not provide insights into the meaning of the latent features. To address this critical issue, we introduce a novel interpretable CAE approach for high-dimensional fluid flow data that maintains the reconstruction quality of conventional CAEs and allows for feature interpretation. Our method can be easily integrated into any existing CAE architecture with minor modifications of the training process. We compare our approach to Proper Orthogonal Decomposition (POD) and two existing methods for interpretable CAEs. We apply all methods to three different experimental turbulent Rayleigh-B\'enard convection datasets with varying complexity. Our results show that the proposed method is lightweight, easy to train, and achieves relative reconstruction performance improvements of up to 6.4% over POD for 64 modes. The relative improvement increases to up to 229.8% as the number of modes decreases. Additionally, our method delivers interpretable features similar to those of POD and is significantly less resource-intensive than existing CAE approaches, using less than 2% of the parameters. These approaches either trade interpretability for reconstruction performance or only provide interpretability to a limited extend.
Authors: Dylan Bouchard, Mohit Singh Chauhan, David Skarbrevik, Viren Bajaj, Zeya Ahmad
Abstract: Large Language Models (LLMs) have been observed to exhibit bias in numerous ways, potentially creating or worsening outcomes for specific groups identified by protected attributes such as sex, race, sexual orientation, or age. To help address this gap, we introduce LangFair, an open-source Python package that aims to equip LLM practitioners with the tools to evaluate bias and fairness risks relevant to their specific use cases. The package offers functionality to easily generate evaluation datasets, comprised of LLM responses to use-case-specific prompts, and subsequently calculate applicable metrics for the practitioner's use case. To guide in metric selection, LangFair offers an actionable decision framework.
Authors: Mingyang Song, Zhaochen Su, Xiaoye Qu, Jiawei Zhou, Yu Cheng
Abstract: Process-level Reward Models (PRMs) are crucial for complex reasoning and decision-making tasks, where each intermediate step plays an important role in the reasoning process. Since language models are prone to various types of errors during the reasoning process, PRMs are required to possess nuanced capabilities for detecting various implicit error types in real-world scenarios. However, current benchmarks primarily focus on step correctness, failing to evaluate PRMs' performance systematically. To address this gap, we introduce PRMBench, a process-level benchmark specifically designed to assess the fine-grained error detection capabilities of PRMs. PRMBench comprises 6,216 carefully designed problems and 83,456 step-level labels, evaluating models across multiple dimensions, including simplicity, soundness, and sensitivity. In our experiments on 15 models, spanning both open-source PRMs and closed-source large language models prompted as critic models, we uncover significant weaknesses in current PRMs. These findings underscore the challenges inherent in process-level evaluation and highlight key directions for future research. We hope PRMBench can be a robust bench for advancing research on PRM evaluation and development.
Authors: Dennis Gross, Helge Spieker
Abstract: Deep reinforcement learning (RL) policies can demonstrate unsafe behaviors and are challenging to interpret. To address these challenges, we combine RL policy model checking--a technique for determining whether RL policies exhibit unsafe behaviors--with co-activation graph analysis--a method that maps neural network inner workings by analyzing neuron activation patterns--to gain insight into the safe RL policy's sequential decision-making. This combination lets us interpret the RL policy's inner workings for safe decision-making. We demonstrate its applicability in various experiments.
Authors: Valery Istomin, Oleg Pereziabov, Ilya Afanasyev
Abstract: This research focuses on developing a method for restoring the topology of digital images of paper documents captured by a camera, using algorithms for detection, segmentation, geometry restoration, and dewarping. Our methodology employs deep learning (DL) for document outline detection, followed by computer vision (CV) to create a topological 2D grid using cubic polynomial interpolation and correct nonlinear distortions by remapping the image. Using classical CV methods makes the document topology restoration process more efficient and faster, as it requires significantly fewer computational resources and memory. We developed a new pipeline for automatic document dewarping and reconstruction, along with a framework and annotated dataset to demonstrate its efficiency. Our experiments confirm the promise of our methodology and its superiority over existing benchmarks (including mobile apps and popular DL solutions, such as RectiNet, DocGeoNet, and DocTr++) both visually and in terms of document readability via Optical Character Recognition (OCR) and geometry restoration metrics. This paves the way for creating high-quality digital copies of paper documents and enhancing the efficiency of OCR systems. Project page: https://github.com/HorizonParadox/DRCCBI
Authors: Alhassan Mumuni, Fuseini Mumuni
Abstract: Generative artificial intelligence (AI) systems based on large-scale pretrained foundation models (PFMs) such as vision-language models, large language models (LLMs), diffusion models and vision-language-action (VLA) models have demonstrated the ability to solve complex and truly non-trivial AI problems in a wide variety of domains and contexts. Multimodal large language models (MLLMs), in particular, learn from vast and diverse data sources, allowing rich and nuanced representations of the world and, thereby, providing extensive capabilities, including the ability to reason, engage in meaningful dialog; collaborate with humans and other agents to jointly solve complex problems; and understand social and emotional aspects of humans. Despite this impressive feat, the cognitive abilities of state-of-the-art LLMs trained on large-scale datasets are still superficial and brittle. Consequently, generic LLMs are severely limited in their generalist capabilities. A number of foundational problems -- embodiment, symbol grounding, causality and memory -- are required to be addressed for LLMs to attain human-level general intelligence. These concepts are more aligned with human cognition and provide LLMs with inherent human-like cognitive properties that support the realization of physically-plausible, semantically meaningful, flexible and more generalizable knowledge and intelligence. In this work, we discuss the aforementioned foundational issues and survey state-of-the art approaches for implementing these concepts in LLMs. Specifically, we discuss how the principles of embodiment, symbol grounding, causality and memory can be leveraged toward the attainment of artificial general intelligence (AGI) in an organic manner.
Authors: Ali Al-Lawati, Jason Lucas, Prasenjit Mitra
Abstract: Large Language Models (LLMs) have demonstrated remarkable performance in various NLP tasks, including semantic parsing, which trans lates natural language into formal code representations. However, the reverse process, translating code into natural language, termed semantic captioning, has received less attention. This task is becoming increasingly important as LLMs are integrated into platforms for code generation, security analysis, and educational purposes. In this paper, we focus on the captioning of SQL query (SQL2Text) to address the critical need for understanding and explaining SQL queries in an era where LLM-generated code poses potential security risks. We repurpose Text2SQL datasets for SQL2Text by introducing an iterative ICL prompt using GPT-4o to generate multiple additional utterances, which enhances the robustness of the datasets for the reverse task. We conduct our experiments using in-context learning (ICL) based on different sample selection methods, emphasizing smaller, more computationally efficient LLMs. Our findings demonstrate that leveraging the inherent graph properties of SQL for ICL sample selection significantly outperforms random selection by up to 39% on BLEU score and provides better results than alternative methods. Dataset and codes are published: \url{https://github.com/aliwister/ast-icl}.
Authors: Jack Boylan, Chris Hokamp, Demian Gholipour Ghalandari
Abstract: We introduce GLiREL (Generalist Lightweight model for zero-shot Relation Extraction), an efficient architecture and training paradigm for zero-shot relation classification. Inspired by recent advancements in zero-shot named entity recognition, this work presents an approach to efficiently and accurately predict zero-shot relationship labels between multiple entities in a single forward pass. Experiments using the FewRel and WikiZSL benchmarks demonstrate that our approach achieves state-of-the-art results on the zero-shot relation classification task. In addition, we contribute a protocol for synthetically-generating datasets with diverse relation labels.
Authors: Holger Severin Bovbjerg (Aalborg University), Jan {\O}stergaard (Aalborg University), Jesper Jensen (Aalborg University, Oticon A/S), Zheng-Hua Tan (Aalborg University)
Abstract: Target-Speaker Voice Activity Detection (TS-VAD) is the task of detecting the presence of speech from a known target-speaker in an audio frame. Recently, deep neural network-based models have shown good performance in this task. However, training these models requires extensive labelled data, which is costly and time-consuming to obtain, particularly if generalization to unseen environments is crucial. To mitigate this, we propose a causal, Self-Supervised Learning (SSL) pretraining framework, called Denoising Autoregressive Predictive Coding (DN-APC), to enhance TS-VAD performance in noisy conditions. We also explore various speaker conditioning methods and evaluate their performance under different noisy conditions. Our experiments show that DN-APC improves performance in noisy conditions, with a general improvement of approx. 2% in both seen and unseen noise. Additionally, we find that FiLM conditioning provides the best overall performance. Representation analysis via tSNE plots reveals robust initial representations of speech and non-speech from pretraining. This underscores the effectiveness of SSL pretraining in improving the robustness and performance of TS-VAD models in noisy environments.
Authors: Dennis Gross
Abstract: In this paper, we propose a novel approach for verifying the compliance of turn-based multi-agent reinforcement learning (TMARL) agents with complex requirements in stochastic multiplayer games. Our method overcomes the limitations of existing verification approaches, which are inadequate for dealing with TMARL agents and not scalable to large games with multiple agents. Our approach relies on tight integration of TMARL and a verification technique referred to as model checking. We demonstrate the effectiveness and scalability of our technique through experiments in different types of environments. Our experiments show that our method is suited to verify TMARL agents and scales better than naive monolithic model checking.
Authors: Xiaoxiao He, Haizhou Shi, Ligong Han, Chaowei Tan, Bo Liu, Zihao Xu, Meng Ye, Leon Axel, Kang Li, Dimitris Metaxas
Abstract: Cardiovascular disease (CVD) and cardiac dyssynchrony are major public health problems in the United States. Precise cardiac image segmentation is crucial for extracting quantitative measures that help categorize cardiac dyssynchrony. However, achieving high accuracy often depends on centralizing large datasets from different hospitals, which can be challenging due to privacy concerns. To solve this problem, Federated Learning (FL) is proposed to enable decentralized model training on such data without exchanging sensitive information. However, bandwidth limitations and data heterogeneity remain as significant challenges in conventional FL algorithms. In this paper, we propose a novel efficient and adaptive federate learning method for cardiac segmentation that improves model performance while reducing the bandwidth requirement. Our method leverages the low-rank adaptation (LoRA) to regularize model weight update and reduce communication overhead. We also propose a \mymethod{} aggregation technique to address data heterogeneity among clients. This technique adaptively penalizes the aggregated weights from different clients by comparing the validation accuracy in each client, allowing better generalization performance and fast local adaptation. In-client and cross-client evaluations on public cardiac MR datasets demonstrate the superiority of our method over other LoRA-based federate learning approaches.
Authors: Yuhui Zhang, Yuchang Su, Yiming Liu, Xiaohan Wang, James Burgess, Elaine Sui, Chenyu Wang, Josiah Aklilu, Alejandro Lozano, Anjiang Wei, Ludwig Schmidt, Serena Yeung-Levy
Abstract: The rapid development of vision language models (VLMs) demands rigorous and reliable evaluation. However, current visual question answering (VQA) benchmarks often depend on open-ended questions, making accurate evaluation difficult due to the variability in natural language responses. To address this, we introduce AutoConverter, an agentic framework that automatically converts these open-ended questions into multiple-choice format, enabling objective evaluation while reducing the costly question creation process. Our experiments demonstrate that AutoConverter can generate correct and challenging multiple-choice questions, with VLMs demonstrating consistently similar or lower accuracy on these questions compared to human-created ones. Using AutoConverter, we construct VMCBench, a benchmark created by transforming 20 existing VQA datasets into a unified multiple-choice format, totaling 9,018 questions. We comprehensively evaluate 33 state-of-the-art VLMs on VMCBench, setting a new standard for scalable, consistent, and reproducible VLM evaluation.
Authors: Beichen Zhang, Yuhong Liu, Xiaoyi Dong, Yuhang Zang, Pan Zhang, Haodong Duan, Yuhang Cao, Dahua Lin, Jiaqi Wang
Abstract: Cutting-edge large language models (LLMs) demonstrate promising performance in solving complex math problems with a divide-and-conquer pipeline and the assistance of in-context learning (ICL) examples. However, their potential for improvement is limited by two critical problems within their ICL examples: granularity-mismatch and the ensuing negative-effect noise problem. Specifically, the LLMs are capable of the dividing process yet mostly failed by inaccurate reasoning within a few conquer steps, while the ICL examples retrieved in question-grained sometimes lack relevant steps for a specific challenging reasoning step. Further, this disconnect may hinder the correct reasoning due to its irrelevance. To this end, we focus on improving the reasoning quality within each step and present BoostStep. BoostStep aligns the granularity between the retrieving and reasoning on step grained, and provides highly related ICL examples for each reasoning step with a novel `first-try' strategy. BoostStep provides more relevant examples than the coarse question-grained strategy, enhancing the model reasoning quality within each step steadily. BoostStep is a general and robust reasoning-enhancing method that not only improves standalone reasoning performance but also integrates seamlessly with Monte Carlo Tree Search methods (MCTS) to refine both candidate generation and decision-making. Quantitatively, it improves GPT-4o and Qwen2.5-Math-72B by 3.6\% and 2.0\% respectively on various mathematical benchmarks, and 7.5\% gain combined with MCTS.
Authors: Guoxuan Chen, Lianghao Xia, Chao Huang
Abstract: Graph neural networks (GNNs) have demonstrated superior performance in collaborative recommendation through their ability to conduct high-order representation smoothing, effectively capturing structural information within users' interaction patterns. However, existing GNN paradigms face significant challenges in scalability and robustness when handling large-scale, noisy, and real-world datasets. To address these challenges, we present LightGNN, a lightweight and distillation-based GNN pruning framework designed to substantially reduce model complexity while preserving essential collaboration modeling capabilities. Our LightGNN framework introduces a computationally efficient pruning module that adaptively identifies and removes redundant edges and embedding entries for model compression. The framework is guided by a resource-friendly hierarchical knowledge distillation objective, whose intermediate layer augments the observed graph to maintain performance, particularly in high-rate compression scenarios. Extensive experiments on public datasets demonstrate LightGNN's effectiveness, significantly improving both computational efficiency and recommendation accuracy. Notably, LightGNN achieves an 80% reduction in edge count and 90% reduction in embedding entries while maintaining performance comparable to more complex state-of-the-art baselines. The implementation of our LightGNN framework is available at the github repository: https://github.com/HKUDS/LightGNN.
Authors: Alexander Korotin, Vladimir V'yugin, Evgeny Burnaev
Abstract: In this paper we extend the setting of the online prediction with expert advice to function-valued forecasts. At each step of the online game several experts predict a function, and the learner has to efficiently aggregate these functional forecasts into a single forecast. We adapt basic mixable (and exponentially concave) loss functions to compare functional predictions and prove that these adaptations are also mixable (exp-concave). We call this phenomenon mixability (exp-concavity) of integral loss functions. As an application of our main result, we prove that various loss functions used for probabilistic forecasting are mixable (exp-concave). The considered losses include Sliced Continuous Ranked Probability Score, Energy-Based Distance, Optimal Transport Costs and Sliced Wasserstein-2 distance, Beta-2 and Kullback-Leibler divergences, Characteristic function and Maximum Mean Discrepancies.
Authors: Yongan Zhang, Yonggan Fu, Weiwen Jiang, Chaojian Li, Haoran You, Meng Li, Vikas Chandra, Yingyan Celine Lin
Abstract: Powerful yet complex deep neural networks (DNNs) have fueled a booming demand for efficient DNN solutions to bring DNN-powered intelligence into numerous applications. Jointly optimizing the networks and their accelerators are promising in providing optimal performance. However, the great potential of such solutions have yet to be unleashed due to the challenge of simultaneously exploring the vast and entangled, yet different design spaces of the networks and their accelerators. To this end, we propose DNA, a Differentiable Network-Accelerator co-search framework for automatically searching for matched networks and accelerators to maximize both the task accuracy and acceleration efficiency. Specifically, DNA integrates two enablers: (1) a generic design space for DNN accelerators that is applicable to both FPGA- and ASIC-based DNN accelerators and compatible with DNN frameworks such as PyTorch to enable algorithmic exploration for more efficient DNNs and their accelerators; and (2) a joint DNN network and accelerator co-search algorithm that enables simultaneously searching for optimal DNN structures and their accelerators' micro-architectures and mapping methods to maximize both the task accuracy and acceleration efficiency. Experiments and ablation studies based on FPGA measurements and ASIC synthesis show that the matched networks and accelerators generated by DNA consistently outperform state-of-the-art (SOTA) DNNs and DNN accelerators (e.g., 3.04x better FPS with a 5.46% higher accuracy on ImageNet), while requiring notably reduced search time (up to 1234.3x) over SOTA co-exploration methods, when evaluated over ten SOTA baselines on three datasets. All codes will be released upon acceptance.
Authors: Yonggan Fu, Zhongzhi Yu, Yongan Zhang, Yingyan Celine Lin
Abstract: AlphaGo's astonishing performance has ignited an explosive interest in developing deep reinforcement learning (DRL) for numerous real-world applications, such as intelligent robotics. However, the often prohibitive complexity of DRL stands at the odds with the required real-time control and constrained resources in many DRL applications, limiting the great potential of DRL powered intelligent devices. While substantial efforts have been devoted to compressing other deep learning models, existing works barely touch the surface of compressing DRL. In this work, we first identify that there exists an optimal model size of DRL that can maximize both the test scores and efficiency, motivating the need for task-specific DRL agents. We therefore propose an Auto-Agent-Distiller (A2D) framework, which to our best knowledge is the first neural architecture search (NAS) applied to DRL to automatically search for the optimal DRL agents for various tasks that optimize both the test scores and efficiency. Specifically, we demonstrate that vanilla NAS can easily fail in searching for the optimal agents, due to its resulting high variance in DRL training stability, and then develop a novel distillation mechanism to distill the knowledge from both the teacher agent's actor and critic to stabilize the searching process and improve the searched agents' optimality. Extensive experiments and ablation studies consistently validate our findings and the advantages and general applicability of our A2D, outperforming manually designed DRL in both the test scores and efficiency. All the codes will be released upon acceptance.
Authors: Yonggan Fu, Han Guo, Meng Li, Xin Yang, Yining Ding, Vikas Chandra, Yingyan Celine Lin
Abstract: Low-precision deep neural network (DNN) training has gained tremendous attention as reducing precision is one of the most effective knobs for boosting DNNs' training time/energy efficiency. In this paper, we attempt to explore low-precision training from a new perspective as inspired by recent findings in understanding DNN training: we conjecture that DNNs' precision might have a similar effect as the learning rate during DNN training, and advocate dynamic precision along the training trajectory for further boosting the time/energy efficiency of DNN training. Specifically, we propose Cyclic Precision Training (CPT) to cyclically vary the precision between two boundary values which can be identified using a simple precision range test within the first few training epochs. Extensive simulations and ablation studies on five datasets and eleven models demonstrate that CPT's effectiveness is consistent across various models/tasks (including classification and language modeling). Furthermore, through experiments and visualization we show that CPT helps to (1) converge to a wider minima with a lower generalization error and (2) reduce training variance which we believe opens up a new design knob for simultaneously improving the optimization and efficiency of DNN training. Our codes are available at: https://github.com/RICE-EIC/CPT.
Authors: Yonggan Fu, Zhongzhi Yu, Yongan Zhang, Yifan Jiang, Chaojian Li, Yongyuan Liang, Mingchao Jiang, Zhangyang Wang, Yingyan Celine Lin
Abstract: The promise of Deep Neural Network (DNN) powered Internet of Thing (IoT) devices has motivated a tremendous demand for automated solutions to enable fast development and deployment of efficient (1) DNNs equipped with instantaneous accuracy-efficiency trade-off capability to accommodate the time-varying resources at IoT devices and (2) dataflows to optimize DNNs' execution efficiency on different devices. Therefore, we propose InstantNet to automatically generate and deploy instantaneously switchable-precision networks which operate at variable bit-widths. Extensive experiments show that the proposed InstantNet consistently outperforms state-of-the-art designs.
Authors: Yonggan Fu, Yongan Zhang, Yang Zhang, David Cox, Yingyan Celine Lin
Abstract: While maximizing deep neural networks' (DNNs') acceleration efficiency requires a joint search/design of three different yet highly coupled aspects, including the networks, bitwidths, and accelerators, the challenges associated with such a joint search have not yet been fully understood and addressed. The key challenges include (1) the dilemma of whether to explode the memory consumption due to the huge joint space or achieve sub-optimal designs, (2) the discrete nature of the accelerator design space that is coupled yet different from that of the networks and bitwidths, and (3) the chicken and egg problem associated with network-accelerator co-search, i.e., co-search requires operation-wise hardware cost, which is lacking during search as the optimal accelerator depending on the whole network is still unknown during search. To tackle these daunting challenges towards optimal and fast development of DNN accelerators, we propose a framework dubbed Auto-NBA to enable jointly searching for the Networks, Bitwidths, and Accelerators, by efficiently localizing the optimal design within the huge joint design space for each target dataset and acceleration specification. Our Auto-NBA integrates a heterogeneous sampling strategy to achieve unbiased search with constant memory consumption, and a novel joint-search pipeline equipped with a generic differentiable accelerator search engine. Extensive experiments and ablation studies validate that both Auto-NBA generated networks and accelerators consistently outperform state-of-the-art designs (including co-search/exploration techniques, hardware-aware NAS methods, and DNN accelerators), in terms of search time, task accuracy, and accelerator efficiency. Our codes are available at: https://github.com/RICE-EIC/Auto-NBA.
Authors: Yonggan Fu, Yongan Zhang, Chaojian Li, Zhongzhi Yu, Yingyan Celine Lin
Abstract: Driven by the explosive interest in applying deep reinforcement learning (DRL) agents to numerous real-time control and decision-making applications, there has been a growing demand to deploy DRL agents to empower daily-life intelligent devices, while the prohibitive complexity of DRL stands at odds with limited on-device resources. In this work, we propose an Automated Agent Accelerator Co-Search (A3C-S) framework, which to our best knowledge is the first to automatically co-search the optimally matched DRL agents and accelerators that maximize both test scores and hardware efficiency. Extensive experiments consistently validate the superiority of our A3C-S over state-of-the-art techniques.
Authors: Jinyung Hong, Theodore P. Pavlic
Abstract: Fruit flies are established model systems for studying olfactory learning as they will readily learn to associate odors with both electric shock or sugar rewards. The mechanisms of the insect brain apparently responsible for odor learning form a relatively shallow neuronal architecture. Olfactory inputs are received by the antennal lobe (AL) of the brain, which produces an encoding of each odor mixture across ~50 sub-units known as glomeruli. Each of these glomeruli then projects its component of this feature vector to several of ~2000 so-called Kenyon Cells (KCs) in a region of the brain known as the mushroom body (MB). Fly responses to odors are generated by small downstream neutrophils that decode the higher-order representation from the MB. Research has shown that there is no recognizable pattern in the glomeruli--KC connections (and thus the particular higher-order representations); they are akin to fingerprints--even isogenic flies have different projections. Leveraging insights from this architecture, we propose KCNet, a single-hidden-layer neural network that contains sparse, randomized, binary weights between the input layer and the hidden layer and analytically learned weights between the hidden layer and the output layer. Furthermore, we also propose a dynamic optimization algorithm that enables the KCNet to increase performance beyond its structural limits by searching for a more efficient set of inputs. For odorant-perception tasks that predict the perceptual properties of an odorant, we show that KCNet outperforms existing data-driven approaches, such as XGBoost. For image classification tasks, KCNet achieves reasonable performance on benchmark datasets (MNIST, Fashion-MNIST, and EMNIST) without any data-augmentation methods or convolutional layers and shows a particularly fast running time.
Authors: Yonggan Fu, Yang Zhao, Qixuan Yu, Chaojian Li, Yingyan Celine Lin
Abstract: The recent breakthroughs of deep neural networks (DNNs) and the advent of billions of Internet of Things (IoT) devices have excited an explosive demand for intelligent IoT devices equipped with domain-specific DNN accelerators. However, the deployment of DNN accelerator enabled intelligent functionality into real-world IoT devices still remains particularly challenging. First, powerful DNNs often come at prohibitive complexities, whereas IoT devices often suffer from stringent resource constraints. Second, while DNNs are vulnerable to adversarial attacks especially on IoT devices exposed to complex real-world environments, many IoT applications require strict security. Existing DNN accelerators mostly tackle only one of the two aforementioned challenges (i.e., efficiency or adversarial robustness) while neglecting or even sacrificing the other. To this end, we propose a 2-in-1 Accelerator, an integrated algorithm-accelerator co-design framework aiming at winning both the adversarial robustness and efficiency of DNN accelerators. Specifically, we first propose a Random Precision Switch (RPS) algorithm that can effectively defend DNNs against adversarial attacks by enabling random DNN quantization as an in-situ model switch. Furthermore, we propose a new precision-scalable accelerator featuring (1) a new precision-scalable MAC unit architecture which spatially tiles the temporal MAC units to boost both the achievable efficiency and flexibility and (2) a systematically optimized dataflow that is searched by our generic accelerator optimizer. Extensive experiments and ablation studies validate that our 2-in-1 Accelerator can not only aggressively boost both the adversarial robustness and efficiency of DNN accelerators under various attacks, but also naturally support instantaneous robustness-efficiency trade-offs adapting to varied resources without the necessity of DNN retraining.
Authors: Yonggan Fu, Qixuan Yu, Yang Zhang, Shang Wu, Xu Ouyang, David Cox, Yingyan Celine Lin
Abstract: Deep Neural Networks (DNNs) are known to be vulnerable to adversarial attacks, i.e., an imperceptible perturbation to the input can mislead DNNs trained on clean images into making erroneous predictions. To tackle this, adversarial training is currently the most effective defense method, by augmenting the training set with adversarial samples generated on the fly. Interestingly, we discover for the first time that there exist subnetworks with inborn robustness, matching or surpassing the robust accuracy of the adversarially trained networks with comparable model sizes, within randomly initialized networks without any model training, indicating that adversarial training on model weights is not indispensable towards adversarial robustness. We name such subnetworks Robust Scratch Tickets (RSTs), which are also by nature efficient. Distinct from the popular lottery ticket hypothesis, neither the original dense networks nor the identified RSTs need to be trained. To validate and understand this fascinating finding, we further conduct extensive experiments to study the existence and properties of RSTs under different models, datasets, sparsity patterns, and attacks, drawing insights regarding the relationship between DNNs' robustness and their initialization/overparameterization. Furthermore, we identify the poor adversarial transferability between RSTs of different sparsity ratios drawn from the same randomly initialized dense network, and propose a Random RST Switch (R2S) technique, which randomly switches between different RSTs, as a novel defense method built on top of RSTs. We believe our findings about RSTs have opened up a new perspective to study model robustness and extend the lottery ticket hypothesis.
Authors: Yonggan Fu, Haichuan Yang, Jiayi Yuan, Meng Li, Cheng Wan, Raghuraman Krishnamoorthi, Vikas Chandra, Yingyan Celine Lin
Abstract: Efficient deep neural network (DNN) models equipped with compact operators (e.g., depthwise convolutions) have shown great potential in reducing DNNs' theoretical complexity (e.g., the total number of weights/operations) while maintaining a decent model accuracy. However, existing efficient DNNs are still limited in fulfilling their promise in boosting real-hardware efficiency, due to their commonly adopted compact operators' low hardware utilization. In this work, we open up a new compression paradigm for developing real-hardware efficient DNNs, leading to boosted hardware efficiency while maintaining model accuracy. Interestingly, we observe that while some DNN layers' activation functions help DNNs' training optimization and achievable accuracy, they can be properly removed after training without compromising the model accuracy. Inspired by this observation, we propose a framework dubbed DepthShrinker, which develops hardware-friendly compact networks via shrinking the basic building blocks of existing efficient DNNs that feature irregular computation patterns into dense ones with much improved hardware utilization and thus real-hardware efficiency. Excitingly, our DepthShrinker framework delivers hardware-friendly compact networks that outperform both state-of-the-art efficient DNNs and compression techniques, e.g., a 3.06% higher accuracy and 1.53$\times$ throughput on Tesla V100 over SOTA channel-wise pruning method MetaPruning. Our codes are available at: https://github.com/facebookresearch/DepthShrinker.
Authors: Yonggan Fu, Yang Zhang, Kaizhi Qian, Zhifan Ye, Zhongzhi Yu, Cheng-I Lai, Yingyan Celine Lin
Abstract: Self-supervised learning (SSL) for rich speech representations has achieved empirical success in low-resource Automatic Speech Recognition (ASR) and other speech processing tasks, which can mitigate the necessity of a large amount of transcribed speech and thus has driven a growing demand for on-device ASR and other speech processing. However, advanced speech SSL models have become increasingly large, which contradicts the limited on-device resources. This gap could be more severe in multilingual/multitask scenarios requiring simultaneously recognizing multiple languages or executing multiple speech processing tasks. Additionally, strongly overparameterized speech SSL models tend to suffer from overfitting when being finetuned on low-resource speech corpus. This work aims to enhance the practical usage of speech SSL models towards a win-win in both enhanced efficiency and alleviated overfitting via our proposed S$^3$-Router framework, which for the first time discovers that simply discarding no more than 10\% of model weights via only finetuning model connections of speech SSL models can achieve better accuracy over standard weight finetuning on downstream speech processing tasks. More importantly, S$^3$-Router can serve as an all-in-one technique to enable (1) a new finetuning scheme, (2) an efficient multilingual/multitask solution, (3) a state-of-the-art ASR pruning technique, and (4) a new tool to quantitatively analyze the learned speech representation. We believe S$^3$-Router has provided a new perspective for practical deployment of speech SSL models. Our codes are available at: https://github.com/GATECH-EIC/S3-Router.
Authors: Jing Yang, Ruibo Wang, Yijun Song, Jihong Li
Abstract: In the task of comparing two classification algorithms, the widely-used McNemar's test aims to infer the presence of a significant difference between the error rates of the two classification algorithms. However, the power of the conventional McNemar's test is usually unpromising because the hold-out (HO) method in the test merely uses a single train-validation split that usually produces a highly varied estimation of the error rates. In contrast, a cross-validation (CV) method repeats the HO method in multiple times and produces a stable estimation. Therefore, a CV method has a great advantage to improve the power of McNemar's test. Among all types of CV methods, a block-regularized 5$\times$2 CV (BCV) has been shown in many previous studies to be superior to the other CV methods in the comparison task of algorithms because the 5$\times$2 BCV can produce a high-quality estimator of the error rate by regularizing the numbers of overlapping records between all training sets. In this study, we compress the 10 correlated contingency tables in the 5$\times$2 BCV to form an effective contingency table. Then, we define a 5$\times$2 BCV McNemar's test on the basis of the effective contingency table. We demonstrate the reasonable type I error and the promising power of the proposed 5$\times$2 BCV McNemar's test on multiple simulated and real-world data sets.
Authors: Jing Liu, Sihan Chen, Xingjian He, Longteng Guo, Xinxin Zhu, Weining Wang, Jinhui Tang
Abstract: In this paper, we propose a Vision-Audio-Language Omni-peRception pretraining model (VALOR) for multi-modal understanding and generation. Different from widely-studied vision-language pretraining models, VALOR jointly models relationships of vision, audio and language in an end-to-end manner. It contains three separate encoders for single modality representations, and a decoder for multimodal conditional text generation. We design two pretext tasks to pretrain VALOR model, including Multimodal Grouping Alignment (MGA) and Multimodal Grouping Captioning (MGC). MGA projects vision, language and audio to the same common space, building vision-language, audio-language and audiovisual-language alignment simultaneously. MGC learns how to generate text tokens in conditions of vision, audio or their both. To promote vision-audio-language pretraining research, we construct a large-scale high-quality tri-modality dataset named VALOR-1M, which contains 1M audiable videos with human annotated audiovisual captions. Extensive experiments show that VALOR can learn strong multimodal correlations and be generalized to various downstream tasks (e.g., retrieval, captioning and question answering), with different input modalities (e.g., vision-language, audio-language and audiovisual-language). VALOR achieves new state-of-the-art performances on series of public cross-modality benchmarks. Code and data are available at project page https://casia-iva-group.github.io/projects/VALOR.
Authors: Yonggan Fu, Ye Yuan, Shang Wu, Jiayi Yuan, Yingyan Celine Lin
Abstract: Transfer learning leverages feature representations of deep neural networks (DNNs) pretrained on source tasks with rich data to empower effective finetuning on downstream tasks. However, the pretrained models are often prohibitively large for delivering generalizable representations, which limits their deployment on edge devices with constrained resources. To close this gap, we propose a new transfer learning pipeline, which leverages our finding that robust tickets can transfer better, i.e., subnetworks drawn with properly induced adversarial robustness can win better transferability over vanilla lottery ticket subnetworks. Extensive experiments and ablation studies validate that our proposed transfer learning pipeline can achieve enhanced accuracy-sparsity trade-offs across both diverse downstream tasks and sparsity patterns, further enriching the lottery ticket hypothesis.
Authors: Hrushikesh Mhaskar
Abstract: For the past 30 years or so, machine learning has stimulated a great deal of research in the study of approximation capabilities (expressive power) of a multitude of processes, such as approximation by shallow or deep neural networks, radial basis function networks, and a variety of kernel based methods. Motivated by applications such as invariant learning, transfer learning, and synthetic aperture radar imaging, we initiate in this paper a general approach to study the approximation capabilities of kernel based networks using non-symmetric kernels. While singular value decomposition is a natural instinct to study such kernels, we consider a more general approach to include the use of a family of kernels, such as generalized translation networks (which include neural networks and translation invariant kernels as special cases) and rotated zonal function kernels. Naturally, unlike traditional kernel based approximation, we cannot require the kernels to be positive definite. In particular, we obtain estimates on the accuracy of uniform approximation of functions in a Sobolev class by ReLU$^r$ networks when $r$ is not necessarily an integer. Our general results apply to the approximation of functions with small smoothness compared to the dimension of the input space.
Authors: Feng-Lei Fan, Ze-Yu Li, Huan Xiong, Tieyong Zeng
Abstract: In this work, beyond width and depth, we augment a neural network with a new dimension called height by intra-linking neurons in the same layer to create an intra-layer hierarchy, which gives rise to the notion of height. We call a neural network characterized by width, depth, and height a 3D network. To put a 3D network in perspective, we theoretically and empirically investigate the expressivity of height. We show via bound estimation and explicit construction that given the same number of neurons and parameters, a 3D ReLU network of width $W$, depth $K$, and height $H$ has greater expressive power than a 2D network of width $H\times W$ and depth $K$, \textit{i.e.}, $\mathcal{O}((2^H-1)W)^K)$ vs $\mathcal{O}((HW)^K)$, in terms of generating more pieces in a piecewise linear function. Next, through approximation rate analysis, we show that by introducing intra-layer links into networks, a ReLU network of width $\mathcal{O}(W)$ and depth $\mathcal{O}(K)$ can approximate polynomials in $[0,1]^d$ with error $\mathcal{O}\left(2^{-2WK}\right)$, which improves $\mathcal{O}\left(W^{-K}\right)$ and $\mathcal{O}\left(2^{-K}\right)$ for fixed width networks. Lastly, numerical experiments on 5 synthetic datasets, 15 tabular datasets, and 3 image benchmarks verify that 3D networks can deliver competitive regression and classification performance.
Authors: Hao Chen, Thomas Barthel
Abstract: Tensor networks developed in the context of condensed matter physics try to approximate order-$N$ tensors with a reduced number of degrees of freedom that is only polynomial in $N$ and arranged as a network of partially contracted smaller tensors. As we have recently demonstrated in the context of quantum many-body physics, computation costs can be further substantially reduced by imposing constraints on the canonical polyadic (CP) rank of the tensors in such networks [arXiv:2205.15296]. Here, we demonstrate how tree tensor networks (TTN) with CP rank constraints and tensor dropout can be used in machine learning. The approach is found to outperform other tensor-network-based methods in Fashion-MNIST image classification. A low-rank TTN classifier with branching ratio $b=4$ reaches a test set accuracy of 90.3\% with low computation costs. Consisting of mostly linear elements, tensor network classifiers avoid the vanishing gradient problem of deep neural networks. The CP rank constraints have additional advantages: The number of parameters can be decreased and tuned more freely to control overfitting, improve generalization properties, and reduce computation costs. They allow us to employ trees with large branching ratios, substantially improving the representation power.
Authors: Lingjing Kong, Shaoan Xie, Weiran Yao, Yujia Zheng, Guangyi Chen, Petar Stojanov, Victor Akinwande, Kun Zhang
Abstract: Unsupervised domain adaptation is critical to many real-world applications where label information is unavailable in the target domain. In general, without further assumptions, the joint distribution of the features and the label is not identifiable in the target domain. To address this issue, we rely on the property of minimal changes of causal mechanisms across domains to minimize unnecessary influences of distribution shifts. To encode this property, we first formulate the data-generating process using a latent variable model with two partitioned latent subspaces: invariant components whose distributions stay the same across domains and sparse changing components that vary across domains. We further constrain the domain shift to have a restrictive influence on the changing components. Under mild conditions, we show that the latent variables are partially identifiable, from which it follows that the joint distribution of data and labels in the target domain is also identifiable. Given the theoretical insights, we propose a practical domain adaptation framework called iMSDA. Extensive experimental results reveal that iMSDA outperforms state-of-the-art domain adaptation algorithms on benchmark datasets, demonstrating the effectiveness of our framework.
Authors: Mengting Ai, Tianxin Wei, Yifan Chen, Zeming Guo, Jingrui He
Abstract: Fine-tuning a pre-trained language model (PLM) emerges as the predominant strategy in many natural language processing applications. However, this process is known to be expensive, especially on edge devices with low computing power. While general approaches (e.g. quantization and distillation) have been widely studied to reduce the compute/memory of PLM fine-tuning, one-shot compression techniques specifically designed for fine-tuning remain largely unexplored. In this paper, we investigate the neural tangent kernel (NTK)--which reveals the gradient descent dynamics of neural networks--of the multilayer perceptrons (MLP) modules in a PLM and propose to coin a lightweight PLM through NTK-approximating MLP fusion. By incorporating NTK into the compression process, MLP Fusion not only preserves the original model's output but also maintains its training dynamics. To achieve this, we reconsider the MLP as a bundle of sub-MLPs and cluster them into a given number of centroids, which can then be restored as a compressed MLP and surprisingly well approximate the NTK of the original PLM. Our approach is applicable to both standard MLP modules and Mixture-of-Experts (MoE) modules in PLMs, demonstrating its scalability and versatility. Additionally, we provide theoretical derivations to demonstrate how the proposed compression preserves the NTK. Extensive experiments of PLM fine-tuning on both natural language understanding and generation tasks are provided to verify the effectiveness of MLP fusion. Our code is available at https://github.com/weitianxin/MLP_Fusion.
Authors: Yongcong Chen, Ting Zeng, Xingyue Chen
Abstract: In this paper, we propose a new approach to building a artificial general intelligence with self awareness, which includes: (1) a new method to implement attention mechanisms; (2) a way to give machines self-demands; (3) how to form a value evaluation system compatible with the network; (4) a way to create the world models; (5) how to realize a top-down, hierarchical thinking decision-making chain; (6) a way to achieve general decision-making and response capabilities; (7) a way for a machine to directly obtain human experience through language. In the paper, we first analyze some of the shortcomings of current LLMs (Large Language Model) and propose ideas for improvement. Then we analyze why our scheme can solve the above problems and provide detailed steps for implementing our scheme. In chapter 4, we have presented a step-by-step mplementation roadmap. And in chapter 5, we have presented a specific implementation demonstration. In chapter 6, we analyze the advantages and disadvantages of our scheme and propose further research directions. In this article, we have put forward how to create genuine artificial general intelligence step by step. It can handle data of all modalities in a unified form and can directly understand the experience that humans already possess through language, thus avoiding the problem that reinforcement learning is required for every decision-making process.
Authors: Yonggan Fu, Yongan Zhang, Zhongzhi Yu, Sixu Li, Zhifan Ye, Chaojian Li, Cheng Wan, Yingyan Celine Lin
Abstract: The remarkable capabilities and intricate nature of Artificial Intelligence (AI) have dramatically escalated the imperative for specialized AI accelerators. Nonetheless, designing these accelerators for various AI workloads remains both labor- and time-intensive. While existing design exploration and automation tools can partially alleviate the need for extensive human involvement, they still demand substantial hardware expertise, posing a barrier to non-experts and stifling AI accelerator development. Motivated by the astonishing potential of large language models (LLMs) for generating high-quality content in response to human language instructions, we embark on this work to examine the possibility of harnessing LLMs to automate AI accelerator design. Through this endeavor, we develop GPT4AIGChip, a framework intended to democratize AI accelerator design by leveraging human natural languages instead of domain-specific languages. Specifically, we first perform an in-depth investigation into LLMs' limitations and capabilities for AI accelerator design, thus aiding our understanding of our current position and garnering insights into LLM-powered automated AI accelerator design. Furthermore, drawing inspiration from the above insights, we develop a framework called GPT4AIGChip, which features an automated demo-augmented prompt-generation pipeline utilizing in-context learning to guide LLMs towards creating high-quality AI accelerator design. To our knowledge, this work is the first to demonstrate an effective pipeline for LLM-powered automated AI accelerator generation. Accordingly, we anticipate that our insights and framework can serve as a catalyst for innovations in next-generation LLM-powered design automation tools.
Authors: Jaiden Fairoze, Sanjam Garg, Somesh Jha, Saeed Mahloujifar, Mohammad Mahmoody, Mingyuan Wang
Abstract: We present a publicly-detectable watermarking scheme for LMs: the detection algorithm contains no secret information, and it is executable by anyone. We embed a publicly-verifiable cryptographic signature into LM output using rejection sampling and prove that this produces unforgeable and distortion-free (i.e., undetectable without access to the public key) text output. We make use of error-correction to overcome periods of low entropy, a barrier for all prior watermarking schemes. We implement our scheme and find that our formal claims are met in practice.
Authors: Zhili He, Yu-Hsing Wang, Jian Zhang
Abstract: AI-based structural design represents a transformative approach that addresses the inefficiencies inherent in traditional structural design practices. This paper innovates the existing AI-based design frameworks from four aspects and proposes Generative AIBIM: an intelligent design pipeline that integrates BIM and generative AI. First, the proposed pipeline not only broadens the application scope of BIM, which aligns with BIM's growing relevance in civil engineering, but also marks a significant supplement to previous methods that relied on CAD drawings. Second, a two-stage generation framework incorporating generative AI (TGAI), inspired by the human drawing process, is designed to simplify the complexity of the design problem. Third, for the AI model in TGAI, this paper pioneers to fuse physical conditions into diffusion models (DMs) to build a novel physics-based conditional diffusion model (PCDM). In contrast to conventional DMs, on the one hand, PCDM directly predicts shear wall drawings to focus on similarity, and on the other hand, PCDM effectively fuses cross-domain information, i.e., design drawings, timesteps, and physical conditions, by integrating well-designed attention modules. Additionally, a new evaluation system including objective and subjective measures is designed to evaluate models' performance, complementing the evaluation system in the traditional methods. The quantitative results demonstrate that PCDM significantly surpasses recent SOTA techniques across both measures. The qualitative results highlight PCDM's superior capabilities in generating high-perceptual-quality drawings adhering to essential design criteria. In addition, benefiting from the fusion of physical conditions, PCDM effectively supports diverse and creative designs tailored to building heights and seismic precautionary intensities, showcasing its unique generation and generalization abilities.
Authors: Ho Fung Tsoi, Vladimir Loncar, Sridhara Dasu, Philip Harris
Abstract: Compact symbolic expressions have been shown to be more efficient than neural network models in terms of resource consumption and inference speed when implemented on custom hardware such as FPGAs, while maintaining comparable accuracy~\cite{tsoi2023symbolic}. These capabilities are highly valuable in environments with stringent computational resource constraints, such as high-energy physics experiments at the CERN Large Hadron Collider. However, finding compact expressions for high-dimensional datasets remains challenging due to the inherent limitations of genetic programming, the search algorithm of most symbolic regression methods. Contrary to genetic programming, the neural network approach to symbolic regression offers scalability to high-dimensional inputs and leverages gradient methods for faster equation searching. Common ways of constraining expression complexity often involve multistage pruning with fine-tuning, which can result in significant performance loss. In this work, we propose $\tt{SymbolNet}$, a neural network approach to symbolic regression specifically designed as a model compression technique, aimed at enabling low-latency inference for high-dimensional inputs on custom hardware such as FPGAs. This framework allows dynamic pruning of model weights, input features, and mathematical operators in a single training process, where both training loss and expression complexity are optimized simultaneously. We introduce a sparsity regularization term for each pruning type, which can adaptively adjust its strength, leading to convergence at a target sparsity ratio. Unlike most existing symbolic regression methods that struggle with datasets containing more than $\mathcal{O}(10)$ inputs, we demonstrate the effectiveness of our model on the LHC jet tagging task (16 inputs), MNIST (784 inputs), and SVHN (3072 inputs).
Authors: Xingyu Qu, Samuel Horvath
Abstract: Model merging offers an efficient way to combine pre-trained neural networks but often suffers from inconsistent performance, especially when merging models with different initializations. We identify the ``vanishing feature'' phenomenon, where input-induced features diminish during propagation through the merged model, degrading performance. Through theoretical and empirical analysis, we reveal that this phenomenon underpins challenges like variance collapse and explains techniques like permutation-based merging, post-merging normalization, etc. We show that existing normalization strategies can be enhanced by precisely targeting the vanishing feature issue. Leveraging these insights, we propose the ``Preserve-First Merging'' (PFM) strategy, which focuses on preserving early-layer features, enabling the merged models, for the first time, to outperform the original models in advanced settings without post-training. Furthermore, we demonstrate that the vanishing feature phenomenon extends to other contexts, such as model pruning. Applying post-pruning normalization to mitigate the issue significantly improves one-shot pruning performance at high sparsity, offering a simple and effective post-pruning solution. The code is available at https://github.com/XingyuQu/VF.
Authors: Yanjun Zhao, Sizhe Dang, Haishan Ye, Guang Dai, Yi Qian, Ivor W. Tsang
Abstract: Fine-tuning large language models (LLMs) with classic first-order optimizers entails prohibitive GPU memory due to the backpropagation process. Recent works have turned to zeroth-order optimizers for fine-tuning, which save substantial memory by using two forward passes. However, these optimizers are plagued by the heterogeneity of parameter curvatures across different dimensions. In this work, we propose HiZOO, a diagonal Hessian informed zeroth-order optimizer which is the first work to leverage the diagonal Hessian to enhance zeroth-order optimizer for fine-tuning LLMs. What's more, HiZOO avoids the expensive memory cost and only increases one forward pass per step. Extensive experiments on various models (350M~66B parameters) indicate that HiZOO improves model convergence, significantly reducing training steps and effectively enhancing model accuracy. Moreover, we visualize the optimization trajectories of HiZOO on test functions, illustrating its effectiveness in handling heterogeneous curvatures. Lastly, we provide theoretical proofs of convergence for HiZOO. Code is publicly available at https://anonymous.4open.science/r/HiZOO27F8.
Authors: Nicola Muca Cirone, Antonio Orvieto, Benjamin Walker, Cristopher Salvi, Terry Lyons
Abstract: Structured state-space models (SSMs) such as S4, stemming from the seminal work of Gu et al., are gaining popularity as effective approaches for modeling sequential data. Deep SSMs demonstrate outstanding performance across a diverse set of domains, at a reduced training and inference cost compared to attention-based transformers. Recent developments show that if the linear recurrence powering SSMs allows for multiplicative interactions between inputs and hidden states (e.g. GateLoop, Mamba, GLA), then the resulting architecture can surpass in both in accuracy and efficiency attention-powered foundation models trained on text, at scales of billion parameters. In this paper, we give theoretical grounding to this recent finding using tools from Rough Path Theory: we show that when random linear recurrences are equipped with simple input-controlled transitions (selectivity mechanism), then the hidden state is provably a low-dimensional projection of a powerful mathematical object called the signature of the input -- capturing non-linear interactions between tokens at distinct timescales. Our theory not only motivates the success of modern selective state-space models such as Mamba but also provides a solid framework to understand the expressive power of future SSM variants.
Authors: Xiongjie Chen, Yunpeng Li
Abstract: Recently, there has been a surge of interest in incorporating neural networks into particle filters, e.g. differentiable particle filters, to perform joint sequential state estimation and model learning for non-linear non-Gaussian state-space models in complex environments. Existing differentiable particle filters are mostly constructed with vanilla neural networks that do not allow density estimation. As a result, they are either restricted to a bootstrap particle filtering framework or employ predefined distribution families (e.g. Gaussian distributions), limiting their performance in more complex real-world scenarios. In this paper we present a differentiable particle filtering framework that uses (conditional) normalizing flows to build its dynamic model, proposal distribution, and measurement model. This not only enables valid probability densities but also allows the proposed method to adaptively learn these modules in a flexible way, without being restricted to predefined distribution families. We derive the theoretical properties of the proposed filters and evaluate the proposed normalizing flow-based differentiable particle filters' performance through a series of numerical experiments.
Authors: Alhassan Mumuni, Fuseini Mumuni
Abstract: Data augmentation is arguably the most important regularization technique commonly used to improve generalization performance of machine learning models. It primarily involves the application of appropriate data transformation operations to create new data samples with desired properties. Despite its effectiveness, the process is often challenging because of the time-consuming trial and error procedures for creating and testing different candidate augmentations and their hyperparameters manually. Automated data augmentation methods aim to automate the process. State-of-the-art approaches typically rely on automated machine learning (AutoML) principles. This work presents a comprehensive survey of AutoML-based data augmentation techniques. We discuss various approaches for accomplishing data augmentation with AutoML, including data manipulation, data integration and data synthesis techniques. We present extensive discussion of techniques for realizing each of the major subtasks of the data augmentation process: search space design, hyperparameter optimization and model evaluation. Finally, we carried out an extensive comparison and analysis of the performance of automated data augmentation techniques and state-of-the-art methods based on classical augmentation approaches. The results show that AutoML methods for data augmentation currently outperform state-of-the-art techniques based on conventional approaches.
Authors: Chen Cecilia Liu, Iryna Gurevych
Abstract: Prior research shows that differences in the early period of neural network training significantly impact the performance of in-distribution (ID) data of tasks. Yet, the implications of early learning dynamics on out-of-distribution (OOD) generalization remain poorly understood, primarily due to the complexities and limitations of existing analytical techniques. In this work, we investigate the relationship between learning dynamics, OOD generalization under covariate shift and the early period of neural network training. We utilize the trace of Fisher Information and sharpness, focusing on gradual unfreezing (i.e., progressively unfreezing parameters during training) as our methodology for investigation. Through a series of empirical experiments, we show that 1) changing the number of trainable parameters during the early period of training via gradual unfreezing can significantly improve OOD results; 2) the trace of Fisher Information and sharpness can be used as indicators for the removal of gradual unfreezing during the early period of training for better OOD generalization. Our experiments on both image and text data show that the early period of training is a general phenomenon that can provide Pareto improvements in ID and OOD performance with minimal complexity. Our work represents a first step towards understanding how early learning dynamics affect neural network OOD generalization under covariate shift and suggests a new avenue to improve and study this problem.
Authors: Payal Varshney, Adriano Lucieri, Christoph Balada, Andreas Dengel, Sheraz Ahmed
Abstract: Trustworthiness is a major prerequisite for the safe application of opaque deep learning models in high-stakes domains like medicine. Understanding the decision-making process not only contributes to fostering trust but might also reveal previously unknown decision criteria of complex models that could advance the state of medical research. The discovery of decision-relevant concepts from black box models is a particularly challenging task. This study proposes Concept Discovery through Latent Diffusion-based Counterfactual Trajectories (CDCT), a novel three-step framework for concept discovery leveraging the superior image synthesis capabilities of diffusion models. In the first step, CDCT uses a Latent Diffusion Model (LDM) to generate a counterfactual trajectory dataset. This dataset is used to derive a disentangled representation of classification-relevant concepts using a Variational Autoencoder (VAE). Finally, a search algorithm is applied to identify relevant concepts in the disentangled latent space. The application of CDCT to a classifier trained on the largest public skin lesion dataset revealed not only the presence of several biases but also meaningful biomarkers. Moreover, the counterfactuals generated within CDCT show better FID scores than those produced by a previously established state-of-the-art method, while being 12 times more resource-efficient. Unsupervised concept discovery holds great potential for the application of trustworthy AI and the further development of human knowledge in various domains. CDCT represents a further step in this direction.
Authors: Shani Goren, Ido Galil, Ran El-Yaniv
Abstract: Deploying deep neural networks for risk-sensitive tasks necessitates an uncertainty estimation mechanism. This paper introduces hierarchical selective classification, extending selective classification to a hierarchical setting. Our approach leverages the inherent structure of class relationships, enabling models to reduce the specificity of their predictions when faced with uncertainty. In this paper, we first formalize hierarchical risk and coverage, and introduce hierarchical risk-coverage curves. Next, we develop algorithms for hierarchical selective classification (which we refer to as "inference rules"), and propose an efficient algorithm that guarantees a target accuracy constraint with high probability. Lastly, we conduct extensive empirical studies on over a thousand ImageNet classifiers, revealing that training regimes such as CLIP, pretraining on ImageNet21k and knowledge distillation boost hierarchical selective performance.
Authors: Rosario Messana, Rui Chen, Andrea Lodi
Abstract: We consider solving a combinatorial optimization problem with an unknown linear constraint using a membership oracle that, given a solution, determines whether it is feasible or infeasible with absolute certainty. The goal of the decision maker is to find the best possible solution subject to a budget on the number of oracle calls. Inspired by active learning based on Support Vector Machines (SVMs), we adapt a classical framework in order to solve the problem by learning and exploiting a surrogate linear constraint. The resulting new framework includes training a linear separator on the labeled points and selecting new points to be labeled, which is achieved by applying a sampling strategy and solving a 0-1 integer linear program. Following the active learning literature, one can consider using SVM as a linear classifier and the information-based sampling strategy known as simple margin. We improve on both sides: we propose an alternative sampling strategy based on mixed-integer quadratic programming and a linear separation method inspired by an algorithm for convex optimization in the oracle model. We conduct experiments on the pure knapsack problem and on a college study plan problem from the literature to show how different linear separation methods and sampling strategies influence the quality of the results in terms of objective value.
Authors: Minh Le, An Nguyen, Huy Nguyen, Trang Nguyen, Trang Pham, Linh Van Ngo, Nhat Ho
Abstract: Exploiting the power of pre-trained models, prompt-based approaches stand out compared to other continual learning solutions in effectively preventing catastrophic forgetting, even with very few learnable parameters and without the need for a memory buffer. While existing prompt-based continual learning methods excel in leveraging prompts for state-of-the-art performance, they often lack a theoretical explanation for the effectiveness of prompting. This paper conducts a theoretical analysis to unravel how prompts bestow such advantages in continual learning, thus offering a new perspective on prompt design. We first show that the attention block of pre-trained models like Vision Transformers inherently encodes a special mixture of experts architecture, characterized by linear experts and quadratic gating score functions. This realization drives us to provide a novel view on prefix tuning, reframing it as the addition of new task-specific experts, thereby inspiring the design of a novel gating mechanism termed Non-linear Residual Gates (NoRGa). Through the incorporation of non-linear activation and residual connection, NoRGa enhances continual learning performance while preserving parameter efficiency. The effectiveness of NoRGa is substantiated both theoretically and empirically across diverse benchmarks and pretraining paradigms. Our code is publicly available at https://github.com/Minhchuyentoancbn/MoE_PromptCL
Authors: Jianqing Song, Jianguo Huang, Wenyu Jiang, Baoming Zhang, Shuangjie Li, Chongjun Wang
Abstract: Graph Neural Networks have achieved remarkable accuracy in semi-supervised node classification tasks. However, these results lack reliable uncertainty estimates. Conformal prediction methods provide a theoretical guarantee for node classification tasks, ensuring that the conformal prediction set contains the ground-truth label with a desired probability (e.g., 95%). In this paper, we empirically show that for each node, aggregating the non-conformity scores of nodes with the same label can improve the efficiency of conformal prediction sets while maintaining valid marginal coverage. This observation motivates us to propose a novel algorithm named Similarity-Navigated Adaptive Prediction Sets (SNAPS), which aggregates the non-conformity scores based on feature similarity and structural neighborhood. The key idea behind SNAPS is that nodes with high feature similarity or direct connections tend to have the same label. By incorporating adaptive similar nodes information, SNAPS can generate compact prediction sets and increase the singleton hit ratio (correct prediction sets of size one). Moreover, we theoretically provide a finite-sample coverage guarantee of SNAPS. Extensive experiments demonstrate the superiority of SNAPS, improving the efficiency of prediction sets and singleton hit ratio while maintaining valid coverage.
Authors: Huidong Liang, Xingchen Wan, Xiaowen Dong
Abstract: We address the problem of optimizing over functions defined on node subsets in a graph. The optimization of such functions is often a non-trivial task given their combinatorial, black-box and expensive-to-evaluate nature. Although various algorithms have been introduced in the literature, most are either task-specific or computationally inefficient and only utilize information about the graph structure without considering the characteristics of the function. To address these limitations, we utilize Bayesian Optimization (BO), a sample-efficient black-box solver, and propose a novel framework for combinatorial optimization on graphs. More specifically, we map each $k$-node subset in the original graph to a node in a new combinatorial graph and adopt a local modeling approach to efficiently traverse the latter graph by progressively sampling its subgraphs using a recursive algorithm. Extensive experiments under both synthetic and real-world setups demonstrate the effectiveness of the proposed BO framework on various types of graphs and optimization tasks, where its behavior is analyzed in detail with ablation studies.
Authors: Chia-Yi Hsu, Yu-Lin Tsai, Chih-Hsun Lin, Pin-Yu Chen, Chia-Mu Yu, Chun-Ying Huang
Abstract: While large language models (LLMs) such as Llama-2 or GPT-4 have shown impressive zero-shot performance, fine-tuning is still necessary to enhance their performance for customized datasets, domain-specific tasks, or other private needs. However, fine-tuning all parameters of LLMs requires significant hardware resources, which can be impractical for typical users. Therefore, parameter-efficient fine-tuning such as LoRA have emerged, allowing users to fine-tune LLMs without the need for considerable computing resources, with little performance degradation compared to fine-tuning all parameters. Unfortunately, recent studies indicate that fine-tuning can increase the risk to the safety of LLMs, even when data does not contain malicious content. To address this challenge, we propose Safe LoRA, a simple one-liner patch to the original LoRA implementation by introducing the projection of LoRA weights from selected layers to the safety-aligned subspace, effectively reducing the safety risks in LLM fine-tuning while maintaining utility. It is worth noting that Safe LoRA is a training-free and data-free approach, as it only requires the knowledge of the weights from the base and aligned LLMs. Our extensive experiments demonstrate that when fine-tuning on purely malicious data, Safe LoRA retains similar safety performance as the original aligned model. Moreover, when the fine-tuning dataset contains a mixture of both benign and malicious data, Safe LoRA mitigates the negative effect made by malicious data while preserving performance on downstream tasks. Our codes are available at \url{https://github.com/IBM/SafeLoRA}.
Authors: Jifei Luo, Hantao Yao, Changsheng Xu
Abstract: Diffusion-based re-ranking is a common method used for retrieving instances by performing similarity propagation in a nearest neighbor graph. However, existing techniques that construct the affinity graph based on pairwise instances can lead to the propagation of misinformation from outliers and other manifolds, resulting in inaccurate results. To overcome this issue, we propose a novel Cluster-Aware Similarity (CAS) diffusion for instance retrieval. The primary concept of CAS is to conduct similarity diffusion within local clusters, which can reduce the influence from other manifolds explicitly. To obtain a symmetrical and smooth similarity matrix, our Bidirectional Similarity Diffusion strategy introduces an inverse constraint term to the optimization objective of local cluster diffusion. Additionally, we have optimized a Neighbor-guided Similarity Smoothing approach to ensure similarity consistency among the local neighbors of each instance. Evaluations in instance retrieval and object re-identification validate the effectiveness of the proposed CAS, our code is publicly available.
Authors: Zhaochun Ren, Zhou Yang, Chenglong Ye, Yufeng Wang, Haizhou Sun, Chao Chen, Xiaofei Zhu, Yunbing Wu, Xiangwen Liao
Abstract: In-context learning (ICL) achieves remarkable performance in various domains such as knowledge acquisition, commonsense reasoning, and semantic understanding. However, its performance significantly deteriorates for emotion detection tasks, especially fine-grained emotion recognition. The underlying reasons for this remain unclear. In this paper, we identify the reasons behind ICL's poor performance from the perspective of prototype theory and propose a method to address this issue. Specifically, we conduct extensive pilot experiments and find that ICL conforms to the prototype theory on fine-grained emotion recognition. Based on this theory, we uncover the following deficiencies in ICL: (1) It relies on prototypes (example-label pairs) that are semantically similar but emotionally inaccurate to predict emotions. (2) It is prone to interference from irrelevant categories, affecting the accuracy and robustness of the predictions. To address these issues, we propose an Emotion Context Learning method (E-ICL) on fine-grained emotion recognition. E-ICL relies on more emotionally accurate prototypes to predict categories by referring to emotionally similar examples with dynamic labels. Simultaneously, E-ICL employs an exclusionary emotion prediction strategy to avoid interference from irrelevant categories, thereby increasing its accuracy and robustness. Note that the entire process is accomplished with the assistance of a plug-and-play emotion auxiliary model, without additional training. Experiments on the fine-grained emotion datasets EDOS, Empathetic-Dialogues, EmpatheticIntent, and GoEmotions show that E-ICL achieves superior emotion prediction performance. Furthermore, even when the emotion auxiliary model used is lower than 10% of the LLMs, E-ICL can still boost the performance of LLMs by over 4% on multiple datasets.
Authors: Nick Hauptvogel, Christian Igel
Abstract: It is common practice to combine deep neural networks into ensembles. These deep ensembles can benefit from the cancellation of errors effect: Errors by ensemble members may average out, leading to better generalization performance than each individual network. Bayesian neural networks learn a posterior distribution over model parameters, and sampling and weighting networks according to this posterior yields an ensemble model referred to as a Bayes ensemble. This study reviews the argument that neither the sampling nor the weighting in Bayes ensembles are particularly well suited for increasing generalization performance, as they do not support the cancellation of errors effect. In contrast, we show that a weighted average of models, where the weights are optimized by minimizing a second-order PAC-Bayesian generalization bound, can improve generalization. It is crucial that the optimization takes correlations between models into account. This can be achieved by minimizing the tandem loss, which requires hold-out data for estimating error correlations. The tandem loss based PAC-Bayesian weighting increases robustness against correlated models and models with lower performance in an ensemble. This allows us to safely add several models from the same learning process to an ensemble, instead of using early-stopping for selecting a single weight configuration. Our experiments provide further evidence that state-of-the-art intricate Bayes ensembles do not outperform simple uniformly weighted deep ensembles in terms of classification accuracy. Additionally, we show that these Bayes ensembles cannot match the performance of deep ensembles weighted by optimizing the tandem loss, which additionally provides nonvacuous rigorous generalization guarantees.
Authors: Tongjun Shi, Shuhao Zhang, Binbin Chen, Bingsheng He
Abstract: Stream Learning (SL) requires models that can quickly adapt to continuously evolving data, posing significant challenges in both computational efficiency and learning accuracy. Effective data selection is critical in SL to ensure a balance between information retention and training efficiency. Traditional rule-based data selection methods struggle to accommodate the dynamic nature of streaming data, highlighting the necessity for innovative solutions that effectively address these challenges. Recent approaches to handling changing data distributions face challenges that limit their effectiveness in fast-paced environments. In response, we propose StreamFP, a novel approach that uniquely employs dynamic, learnable parameters called fingerprints to enhance data selection efficiency and adaptability in stream learning. StreamFP optimizes coreset selection through its unique fingerprint-guided mechanism for efficient training while ensuring robust buffer updates that adaptively respond to data dynamics, setting it apart from existing methods in stream learning. Experimental results demonstrate that StreamFP outperforms state-of-the-art methods by achieving accuracy improvements of 15.99%, 29.65%, and 51.24% compared to baseline models across varying data arrival rates, alongside a training throughput increase of 4.6x.
Authors: Georgia Papacharalampous, Hristos Tyralis, Nikolaos Doulamis, Anastasios Doulamis
Abstract: To facilitate effective decision-making, precipitation datasets should include uncertainty estimates. Quantile regression with machine learning has been proposed for issuing such estimates. Distributional regression offers distinct advantages over quantile regression, including the ability to model intermittency as well as a stronger ability to extrapolate beyond the training data, which is critical for predicting extreme precipitation. Therefore, here, we introduce the concept of distributional regression in precipitation dataset creation, specifically for the spatial prediction task of correcting satellite precipitation products. Building upon this concept, we formulated new ensemble learning methods that can be valuable not only for spatial prediction but also for other prediction problems. These methods exploit conditional zero-adjusted probability distributions estimated with generalized additive models for location, scale and shape (GAMLSS), spline-based GAMLSS and distributional regression forests as well as their ensembles (stacking based on quantile regression and equal-weight averaging). To identify the most effective methods for our specific problem, we compared them to benchmarks using a large, multi-source precipitation dataset. Stacking was shown to be superior to individual methods at most quantile levels when evaluated with the quantile loss function. Moreover, while the relative ranking of the methods varied across different quantile levels, stacking methods, and to a lesser extent mean combiners, exhibited lower variance in their performance across different quantiles compared to individual methods that occasionally ranked extremely low. Overall, a task-specific combination of multiple distributional regression algorithms could yield significant benefits in terms of stability.
Authors: Xuan Liu, Siqi Cai, Qihua Zhou, Song Guo, Ruibin Li, Kaiwei Lin
Abstract: Perturbation-based mechanisms, such as differential privacy, mitigate gradient leakage attacks by introducing noise into the gradients, thereby preventing attackers from reconstructing clients' private data from the leaked gradients. However, can gradient perturbation protection mechanisms truly defend against all gradient leakage attacks? In this paper, we present the first attempt to break the shield of gradient perturbation protection in Federated Learning for the extraction of private information. We focus on common noise distributions, specifically Gaussian and Laplace, and apply our approach to DNN and CNN models. We introduce Mjolnir, a perturbation-resilient gradient leakage attack that is capable of removing perturbations from gradients without requiring additional access to the original model structure or external data. Specifically, we leverage the inherent diffusion properties of gradient perturbation protection to develop a novel diffusion-based gradient denoising model for Mjolnir. By constructing a surrogate client model that captures the structure of perturbed gradients, we obtain crucial gradient data for training the diffusion model. We further utilize the insight that monitoring disturbance levels during the reverse diffusion process can enhance gradient denoising capabilities, allowing Mjolnir to generate gradients that closely approximate the original, unperturbed versions through adaptive sampling steps. Extensive experiments demonstrate that Mjolnir effectively recovers the protected gradients and exposes the Federated Learning process to the threat of gradient leakage, achieving superior performance in gradient denoising and private data recovery.
Authors: Johnny Jingze Li, Vivek Kurien George, Gabriel A. Silva
Abstract: Emergence in machine learning refers to the spontaneous appearance of complex behaviors or capabilities that arise from the scale and structure of training data and model architectures, despite not being explicitly programmed. We introduce a novel yet straightforward neural network initialization scheme that aims at achieving greater potential for emergence. Measuring emergence as a kind of structural nonlinearity, our method adjusts the layer-wise weight scaling factors to achieve higher emergence values. This enhancement is easy to implement, requiring no additional optimization steps for initialization compared to GradInit. We evaluate our approach across various architectures, including MLP and convolutional architectures for image recognition and transformers for machine translation. We demonstrate substantial improvements in both model accuracy and training speed, with and without batch normalization. The simplicity, theoretical innovation, and demonstrable empirical advantages of our method make it a potent enhancement to neural network initialization practices. These results suggest a promising direction for leveraging emergence to improve neural network training methodologies. Code is available at: https://github.com/johnnyjingzeli/EmergenceInit.
Authors: Cheems Wang, Yiqin Lv, Yixiu Mao, Yun Qu, Yi Xu, Xiangyang Ji
Abstract: Meta-learning is a practical learning paradigm to transfer skills across tasks from a few examples. Nevertheless, the existence of task distribution shifts tends to weaken meta-learners' generalization capability, particularly when the training task distribution is naively hand-crafted or based on simple priors that fail to cover critical scenarios sufficiently. Here, we consider explicitly generative modeling task distributions placed over task identifiers and propose robustifying fast adaptation from adversarial training. Our approach, which can be interpreted as a model of a Stackelberg game, not only uncovers the task structure during problem-solving from an explicit generative model but also theoretically increases the adaptation robustness in worst cases. This work has practical implications, particularly in dealing with task distribution shifts in meta-learning, and contributes to theoretical insights in the field. Our method demonstrates its robustness in the presence of task subpopulation shifts and improved performance over SOTA baselines in extensive experiments. The code is available at the project site https://sites.google.com/view/ar-metalearn.
Authors: Heejoon Koo
Abstract: In this paper, we present NECHO v2, a novel framework designed to enhance the predictive accuracy of multimodal sequential patient diagnoses under uncertain missing visit sequences, a common challenge in real clinical settings. Firstly, we modify NECHO, designed in a diagnosis code-centric fashion, to handle uncertain modality representation dominance under the imperfect data. Secondly, we develop a systematic knowledge distillation by employing the modified NECHO as both teacher and student. It encompasses a modality-wise contrastive and hierarchical distillation, transformer representation random distillation, along with other distillations to align representations between teacher and student tightly and effectively. We also propose curriculum learning guided random data erasing within sequences during both training and distillation of the teacher to lightly simulate scenario with missing visit information, thereby fostering effective knowledge transfer. As a result, NECHO v2 verifies itself by showing robust superiority in multimodal sequential diagnosis prediction under both balanced and imbalanced incomplete settings on multimodal healthcare data.
Authors: Wenbin Zhou, Jin-Hong Du
Abstract: The spatial context of single-cell gene expression data is crucial for many downstream analyses, yet often remains inaccessible due to practical and technical limitations, restricting the utility of such datasets. In this paper, we propose a generic representation learning and transfer learning framework dp-VAE, capable of reconstructing the spatial coordinates associated with the provided gene expression data. Central to our approach is a distance-preserving regularizer integrated into the loss function during training, ensuring the model effectively captures and utilizes spatial context signals from reference datasets. During the inference stage, the produced latent representation of the model can be used to reconstruct or impute the spatial context of the provided gene expression by solving a constrained optimization problem. We also explore the theoretical connections between distance-preserving loss, distortion, and the bi-Lipschitz condition within generative models. Finally, we demonstrate the effectiveness of dp-VAE in different tasks involving training robustness, out-of-sample evaluation, and transfer learning inference applications by testing it over 27 publicly available datasets. This underscores its applicability to a wide range of genomics studies that were previously hindered by the absence of spatial data.
Authors: Wangying Yang, Zitao Zheng, Zhizhong Wu, Bo Zhang, Yuanfang Yang
Abstract: This study introduces a pioneering Dynamic Hypergraph Networks (DHCE) model designed to predict future medical diagnoses from electronic health records with enhanced accuracy. The DHCE model innovates by identifying and differentiating acute and chronic diseases within a patient's visit history, constructing dynamic hypergraphs that capture the complex, high-order interactions between diseases. It surpasses traditional recurrent neural networks and graph neural networks by effectively integrating clinical event data, reflected through medical language model-assisted encoding, into a robust patient representation. Through extensive experiments on two benchmark datasets, MIMIC-III and MIMIC-IV, the DHCE model exhibits superior performance, significantly outpacing established baseline models in the precision of sequential diagnosis prediction.
Authors: Yarin Bar, Shalev Shaer, Yaniv Romano
Abstract: We present a novel approach for test-time adaptation via online self-training, consisting of two components. First, we introduce a statistical framework that detects distribution shifts in the classifier's entropy values obtained on a stream of unlabeled samples. Second, we devise an online adaptation mechanism that utilizes the evidence of distribution shifts captured by the detection tool to dynamically update the classifier's parameters. The resulting adaptation process drives the distribution of test entropy values obtained from the self-trained classifier to match those of the source domain, building invariance to distribution shifts. This approach departs from the conventional self-training method, which focuses on minimizing the classifier's entropy. Our approach combines concepts in betting martingales and online learning to form a detection tool capable of quickly reacting to distribution shifts. We then reveal a tight relation between our adaptation scheme and optimal transport, which forms the basis of our novel self-supervised loss. Experimental results demonstrate that our approach improves test-time accuracy under distribution shifts while maintaining accuracy and calibration in their absence, outperforming leading entropy minimization methods across various scenarios.
Authors: Gao Tianci, Dmitriev D. Dmitry, Konstantin A. Neusypin, Yang Bo, Rao Shengren
Abstract: Recent advancements in reinforcement learning (RL) have been fueled by large-scale data and deep neural networks, particularly for high-dimensional and complex tasks. Online RL methods like Proximal Policy Optimization (PPO) are effective in dynamic scenarios but require substantial real-time data, posing challenges in resource-constrained or slow simulation environments. Offline RL addresses this by pre-learning policies from large datasets, though its success depends on the quality and diversity of the data. This work proposes a framework that enhances PPO algorithms by incorporating a diffusion model to generate high-quality virtual trajectories for offline datasets. This approach improves exploration and sample efficiency, leading to significant gains in cumulative rewards, convergence speed, and strategy stability in complex tasks. Our contributions are threefold: we explore the potential of diffusion models in RL, particularly for offline datasets, extend the application of online RL to offline environments, and experimentally validate the performance improvements of PPO with diffusion models. These findings provide new insights and methods for applying RL to high-dimensional, complex tasks. Finally, we open-source our code at https://github.com/TianciGao/DiffPPO
Authors: Anita Eisenb\"urger, Daniel Otten, Anselm Hudde, Frank Hopfgartner
Abstract: Label noise, which refers to the mislabeling of instances in a dataset, can significantly impair classifier performance, increase model complexity, and affect feature selection. While most research has concentrated on deep neural networks for image and text data, this study explores the impact of label noise on gradient-boosted decision trees (GBDTs), the leading algorithm for tabular data. This research fills a gap by examining the robustness of GBDTs to label noise, focusing on adapting two noise detection methods from deep learning for use with GBDTs and introducing a new detection method called Gradients. Additionally, we extend a method initially designed for GBDTs to incorporate relabeling. By using diverse datasets such as Covertype and Breast Cancer, we systematically introduce varying levels of label noise and evaluate the effectiveness of early stopping and noise detection methods in maintaining model performance. Our noise detection methods achieve state-of-the-art results, with a noise detection accuracy above 99% on the Adult dataset across all noise levels. This work enhances the understanding of label noise in GBDTs and provides a foundation for future research in noise detection and correction methods.
Authors: Liang Zhang, Jionghao Lin, John Sabatini, Conrad Borchers, Daniel Weitekamp, Meng Cao, John Hollander, Xiangen Hu, Arthur C. Graesser
Abstract: Learning performance data describe correct and incorrect answers or problem-solving attempts in adaptive learning, such as in intelligent tutoring systems (ITSs). Learning performance data tend to be highly sparse (80\%\(\sim\)90\% missing observations) in most real-world applications due to adaptive item selection. This data sparsity presents challenges to using learner models to effectively predict future performance explore new hypotheses about learning. This article proposes a systematic framework for augmenting learner data to address data sparsity in learning performance data. First, learning performance is represented as a three-dimensional tensor of learners' questions, answers, and attempts, capturing longitudinal knowledge states during learning. Second, a tensor factorization method is used to impute missing values in sparse tensors of collected learner data, thereby grounding the imputation on knowledge tracing tasks that predict missing performance values based on real observations. Third, a module for generating patterns of learning is used. This study contrasts two forms of generative Artificial Intelligence (AI), including Generative Adversarial Networks (GANs) and Generate Pre-Trained Transformers (GPT) to generate data associated with different clusters of learner data. We tested this approach on an adult literacy dataset from AutoTutor lessons developed for Adult Reading Comprehension (ARC). We found that: (1) tensor factorization improved the performance in tracing and predicting knowledge mastery compared with other knowledge tracing techniques without data augmentation, showing higher relative fidelity for this imputation method, and (2) the GAN-based simulation showed greater overall stability and less statistical bias based on a divergence evaluation with varying simulation sample sizes compared to GPT.
Authors: Sukai Huang, Shu-Wei Liu, Nir Lipovetzky, Trevor Cohn
Abstract: While Vision-Language Models (VLMs) are increasingly used to generate reward signals for training embodied agents to follow instructions, our research reveals that agents guided by VLM rewards often underperform compared to those employing only intrinsic (exploration-driven) rewards, contradicting expectations set by recent work. We hypothesize that false positive rewards -- instances where unintended trajectories are incorrectly rewarded -- are more detrimental than false negatives. Our analysis confirms this hypothesis, revealing that the widely used cosine similarity metric is prone to false positive reward estimates. To address this, we introduce BiMI ({Bi}nary {M}utual {I}nformation), a novel reward function designed to mitigate noise. BiMI significantly enhances learning efficiency across diverse and challenging embodied navigation environments. Our findings offer a nuanced understanding of how different types of reward noise impact agent learning and highlight the importance of addressing multimodal reward signal noise when training embodied agents
Authors: Matteo Carnelos, Francesco Pasti, Nicola Bellotto
Abstract: In recent years, there has been a significant interest in developing machine learning algorithms on embedded systems. This is particularly relevant for bare metal devices in Internet of Things, Robotics, and Industrial applications that face limited memory, processing power, and storage, and which require extreme robustness. To address these constraints, we present MicroFlow, an open-source TinyML framework for the deployment of Neural Networks (NNs) on embedded systems using the Rust programming language. The compiler-based inference engine of MicroFlow, coupled with Rust's memory safety, makes it suitable for TinyML applications in critical environments. The proposed framework enables the successful deployment of NNs on highly resource-constrained devices, including bare-metal 8-bit microcontrollers with only 2kB of RAM. Furthermore, MicroFlow is able to use less Flash and RAM memory than other state-of-the-art solutions for deploying NN reference models (i.e. wake-word and person detection), achieving equally accurate but faster inference compared to existing engines on medium-size NNs, and similar performance on bigger ones. The experimental results prove the efficiency and suitability of MicroFlow for the deployment of TinyML models in critical environments where resources are particularly limited.
Authors: Ezra Karger, Houtan Bastani, Chen Yueh-Han, Zachary Jacobs, Danny Halawi, Fred Zhang, Philip E. Tetlock
Abstract: Forecasts of future events are essential inputs into informed decision-making. Machine learning (ML) systems have the potential to deliver forecasts at scale, but there is no framework for evaluating the accuracy of ML systems on a standardized set of forecasting questions. To address this gap, we introduce ForecastBench: a dynamic benchmark that evaluates the accuracy of ML systems on an automatically generated and regularly updated set of 1,000 forecasting questions. To avoid any possibility of data leakage, ForecastBench is comprised solely of questions about future events that have no known answer at the time of submission. We quantify the capabilities of current ML systems by collecting forecasts from expert (human) forecasters, the general public, and LLMs on a random subset of questions from the benchmark ($N=200$). While LLMs have achieved super-human performance on many benchmarks, they perform less well here: expert forecasters outperform the top-performing LLM (p-value $<0.01$). We display system and human scores in a public leaderboard at www.forecastbench.org.
Authors: Swadesh Swain, Shree Singhi
Abstract: Integrated Gradients (IG) is a widely used algorithm for attributing the outputs of a deep neural network to its input features. Due to the absence of closed-form integrals for deep learning models, inaccurate Riemann Sum approximations are used to calculate IG. This often introduces undesirable errors in the form of high levels of noise, leading to false insights in the model's decision-making process. We introduce a framework, RiemannOpt, that minimizes these errors by optimizing the sample point selection for the Riemann Sum. Our algorithm is highly versatile and applicable to IG as well as its derivatives like Blur IG and Guided IG. RiemannOpt achieves up to 20% improvement in Insertion Scores. Additionally, it enables its users to curtail computational costs by up to four folds, thereby making it highly functional for constrained environments.
Authors: Chao Feng, Hongjie Guan, Alberto Huertas Celdr\'an, Jan von der Assen, G\'er\^ome Bovet, Burkhard Stiller
Abstract: Non-Independent and Identically Distributed (non-IID) data in Federated Learning (FL) causes client drift issues, leading to slower convergence and reduced model performance. While existing approaches mitigate this issue in Centralized FL (CFL) using a central server, Decentralized FL (DFL) remains underexplored. In DFL, the absence of a central entity results in nodes accessing a global view of the federation, further intensifying the challenges of non-IID data. Drawing on the entropy pooling algorithm employed in financial contexts to synthesize diverse investment opinions, this work proposes the Federated Entropy Pooling (FedEP) algorithm to mitigate the non-IID challenge in DFL. FedEP leverages Gaussian Mixture Models (GMM) to fit local data distributions, sharing statistical parameters among neighboring nodes to estimate the global distribution. Aggregation weights are determined using the entropy pooling approach between local and global distributions. By sharing only synthetic distribution information, FedEP preserves data privacy while minimizing communication overhead. Experimental results demonstrate that FedEP achieves faster convergence and outperforms state-of-the-art methods in various non-IID settings.
Authors: Yuling Jiao, Huazhen Lin, Yuchen Luo, Jerry Zhijian Yang
Abstract: This paper presents a framework for deep transfer learning, which aims to leverage information from multi-domain upstream data with a large number of samples $n$ to a single-domain downstream task with a considerably smaller number of samples $m$, where $m \ll n$, in order to enhance performance on downstream task. Our framework offers several intriguing features. First, it allows the existence of both shared and domain-specific features across multi-domain data and provides a framework for automatic identification, achieving precise transfer and utilization of information. Second, the framework explicitly identifies upstream features that contribute to downstream tasks, establishing clear relationships between upstream domains and downstream tasks, thereby enhancing interpretability. Error analysis shows that our framework can significantly improve the convergence rate for learning Lipschitz functions in downstream supervised tasks, reducing it from $\tilde{O}(m^{-\frac{1}{2(d+2)}}+n^{-\frac{1}{2(d+2)}})$ ("no transfer") to $\tilde{O}(m^{-\frac{1}{2(d^*+3)}} + n^{-\frac{1}{2(d+2)}})$ ("partial transfer"), and even to $\tilde{O}(m^{-1/2}+n^{-\frac{1}{2(d+2)}})$ ("complete transfer"), where $d^* \ll d$ and $d$ is the dimension of the observed data. Our theoretical findings are supported by empirical experiments on image classification and regression datasets.
Authors: Weikai Li, Ding Wang, Zijian Ding, Atefeh Sohrabizadeh, Zongyue Qin, Jason Cong, Yizhou Sun
Abstract: High-level synthesis (HLS) is a widely used tool in designing Field Programmable Gate Array (FPGA). HLS enables FPGA design with software programming languages by compiling the source code into an FPGA circuit. The source code includes a program (called ``kernel'') and several pragmas that instruct hardware synthesis, such as parallelization, pipeline, etc. While it is relatively easy for software developers to design the program, it heavily relies on hardware knowledge to design the pragmas, posing a big challenge for software developers. Recently, different machine learning algorithms, such as GNNs, have been proposed to automate the pragma design via performance prediction. However, when applying the trained model on new kernels, the significant domain shift often leads to unsatisfactory performance. We propose a more domain-generalizable model structure: a two-level hierarchical Mixture of Experts (MoE), that can be flexibly adapted to any GNN model. Different expert networks can learn to deal with different regions in the representation space, and they can utilize similar patterns between the old kernels and new kernels. In the low-level MoE, we apply MoE on three natural granularities of a program: node, basic block, and graph. The high-level MoE learns to aggregate the three granularities for the final decision. To stably train the hierarchical MoE, we further propose a two-stage training method. Extensive experiments verify the effectiveness of the hierarchical MoE.
Authors: Francisco Erivaldo Fernandes Junior, Antti Oulasvirta
Abstract: Developing a reinforcement learning (RL) agent often involves identifying values for numerous parameters, covering the policy, reward function, environment, and agent-internal architecture. Since these parameters are interrelated in complex ways, optimizing them is a black-box problem that proves especially challenging for nonexperts. Although existing optimization-as-a-service platforms (e.g., Vizier and Optuna) can handle such problems, they are impractical for RL systems, since the need for manual user mapping of each parameter to distinct components makes the effort cumbersome. It also requires understanding of the optimization process, limiting the systems' application beyond the machine learning field and restricting access in areas such as cognitive science, which models human decision-making. To tackle these challenges, the paper presents \name, a flexible low-code platform to optimize any parameter set across an RL system. Available at https://github.com/feferna/AgentForge, it allows an optimization problem to be defined in a few lines of code and handed to any of the interfaced optimizers. With AgentForge, the user can optimize the parameters either individually or jointly. The paper presents an evaluation of its performance for a challenging vision-based RL problem.
Authors: Vivek Singh, Shikha Chaganti, Matthias Siebert, Sowmya Rajesh, Andrei Puiu, Raj Gopalan, Jamie Gramz, Dorin Comaniciu, Ali Kamen
Abstract: Early screening for cancer has proven to improve the survival rate and spare patients from intensive and costly treatments due to late diagnosis. Cancer screening in the healthy population involves an initial risk stratification step to determine the screening method and frequency, primarily to optimize resource allocation by targeting screening towards individuals who draw most benefit. For most screening programs, age and clinical risk factors such as family history are part of the initial risk stratification algorithm. In this paper, we focus on developing a blood marker-based risk stratification approach, which could be used to identify patients with elevated cancer risk to be encouraged for taking a diagnostic test or participate in a screening program. We demonstrate that the combination of simple, widely available blood tests, such as complete blood count and complete metabolic panel, could potentially be used to identify patients at risk for colorectal, liver, and lung cancers with areas under the ROC curve of 0.76, 0.85, 0.78, respectively. Furthermore, we hypothesize that such an approach could not only be used as pre-screening risk assessment for individuals but also as population health management tool, for example to better interrogate the cancer risk in certain sub-populations.
Authors: Xinrui He, Yikun Ban, Jiaru Zou, Tianxin Wei, Curtiss B. Cook, Jingrui He
Abstract: Missing data imputation is a critical challenge in various domains, such as healthcare and finance, where data completeness is vital for accurate analysis. Large language models (LLMs), trained on vast corpora, have shown strong potential in data generation, making them a promising tool for data imputation. However, challenges persist in designing effective prompts for a finetuning-free process and in mitigating the risk of LLM hallucinations. To address these issues, we propose a novel framework, LLM-Forest, which introduces a "forest" of few-shot learning LLM "trees" with confidence-based weighted voting, inspired by ensemble learning (Random Forest). This framework is established on a new concept of bipartite information graphs to identify high-quality relevant neighboring entries with both feature and value granularity. Extensive experiments on 9 real-world datasets demonstrate the effectiveness and efficiency of LLM-Forest.
Authors: Navyansh Mahla, Ganesh Ramakrishnan
Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities across various domains, particularly in task generalization for both text and vision data. While fine-tuning these models can significantly enhance their performance on specific downstream tasks, it often requires high-quality data that cannot be shared due to privacy concerns. Federated Learning (FL) offers a promising solution for collaborative training without direct data sharing. However, many parameter-efficient fine-tuning strategies for LLMs in FL, particularly those based on Low-Rank Adaptation (LoRA), face limitations. In this paper, we critically analyze the convergence and performance guarantees of popular FL frameworks utilizing LoRA, highlighting its suboptimal nature due to constrained subspace learning of low-rank matrices. This limitation hinders effective fine-tuning of LLMs in federated settings. Through rigorous analytical and empirical evaluations, we demonstrate that direct weight averaging outperforms LoRA-based strategies, leading to superior performance for fine-tuned models. Our comprehensive comparison unmasks inefficiencies in LoRA approaches and underscores the advantages of direct weight aggregation. We extend our analysis to low-rank gradient-based optimizers, such as GaLore, used during local training steps. Our findings show that GaLore along with direct-weight aggregation is a more effective approach, outperforming federated LoRA methods like FlexLoRA and FFA-LoRA across both text and image modalities. While privacy remains paramount in FL discourse, our focus is on assessing performance outcomes of federated fine-tuned models and evaluating various FL frameworks from both theoretical and empirical perspectives. Our findings advocate reassessing the reliance on LoRA within FL contexts, paving the way for more efficient training methodologies.
Authors: Yujie Zhao, Jose Efraim Aguilar Escamill, Weyl Lu, Huazheng Wang
Abstract: Reinforcement Learning from Human Feedback (RLHF) has recently surged in popularity, particularly for aligning large language models and other AI systems with human intentions. At its core, RLHF can be viewed as a specialized instance of Preference-based Reinforcement Learning (PbRL), where the preferences specifically originate from human judgments rather than arbitrary evaluators. Despite this connection, most existing approaches in both RLHF and PbRL primarily focus on optimizing a mean reward objective, neglecting scenarios that necessitate risk-awareness, such as AI safety, healthcare, and autonomous driving. These scenarios often operate under a one-episode-reward setting, which makes conventional risk-sensitive objectives inapplicable. To address this, we explore and prove the applicability of two risk-aware objectives to PbRL : nested and static quantile risk objectives. We also introduce Risk-AwarePbRL (RA-PbRL), an algorithm designed to optimize both nested and static objectives. Additionally, we provide a theoretical analysis of the regret upper bounds, demonstrating that they are sublinear with respect to the number of episodes, and present empirical results to support our findings. Our code is available in https://github.com/aguilarjose11/PbRLNeurips.
Authors: Md Abrar Jahin, Md. Akmol Masud, Md Wahiduzzaman Suva, M. F. Mridha, Nilanjan Dey
Abstract: The rapid data surge from the high-luminosity Large Hadron Collider introduces critical computational challenges requiring novel approaches for efficient data processing in particle physics. Quantum machine learning, with its capability to leverage the extensive Hilbert space of quantum hardware, offers a promising solution. However, current quantum graph neural networks (GNNs) lack robustness to noise and are often constrained by fixed symmetry groups, limiting adaptability in complex particle interaction modeling. This paper demonstrates that replacing the Lorentz Group Equivariant Block modules in LorentzNet with a dressed quantum circuit significantly enhances performance despite using nearly 5.5 times fewer parameters. Additionally, quantum circuits effectively replace MLPs by inherently preserving symmetries, with Lorentz symmetry integration ensuring robust handling of relativistic invariance. Our Lorentz-Equivariant Quantum Graph Neural Network (Lorentz-EQGNN) achieved 74.00% test accuracy and an AUC of 87.38% on the Quark-Gluon jet tagging dataset, outperforming the classical and quantum GNNs with a reduced architecture using only 4 qubits. On the Electron-Photon dataset, Lorentz-EQGNN reached 67.00% test accuracy and an AUC of 68.20%, demonstrating competitive results with just 800 training samples. Evaluation of our model on generic MNIST and FashionMNIST datasets confirmed Lorentz-EQGNN's efficiency, achieving 88.10% and 74.80% test accuracy, respectively. Ablation studies validated the impact of quantum components on performance, with notable improvements in background rejection rates over classical counterparts. These results highlight Lorentz-EQGNN's potential for immediate applications in noise-resilient jet tagging, event classification, and broader data-scarce HEP tasks.
Authors: Ruiquan Huang, Yingbin Liang, Jing Yang
Abstract: Distributionally robust offline reinforcement learning (RL) aims to find a policy that performs the best under the worst environment within an uncertainty set using an offline dataset collected from a nominal model. While recent advances in robust RL focus on Markov decision processes (MDPs), robust non-Markovian RL is limited to planning problem where the transitions in the uncertainty set are known. In this paper, we study the learning problem of robust offline non-Markovian RL. Specifically, when the nominal model admits a low-rank structure, we propose a new algorithm, featuring a novel dataset distillation and a lower confidence bound (LCB) design for robust values under different types of the uncertainty set. We also derive new dual forms for these robust values in non-Markovian RL, making our algorithm more amenable to practical implementation. By further introducing a novel type-I concentrability coefficient tailored for offline low-rank non-Markovian decision processes, we prove that our algorithm can find an $\epsilon$-optimal robust policy using $O(1/\epsilon^2)$ offline samples. Moreover, we extend our algorithm to the case when the nominal model does not have specific structure. With a new type-II concentrability coefficient, the extended algorithm also enjoys polynomial sample efficiency under all different types of the uncertainty set.
Authors: Mohammad Saiful Islam, Mohamed Sami Rakha, William Pourmajidi, Janakan Sivaloganathan, John Steinbacher, Andriy Miranskyy
Abstract: As Large-Scale Cloud Systems (LCS) become increasingly complex, effective anomaly detection is critical for ensuring system reliability and performance. However, there is a shortage of large-scale, real-world datasets available for benchmarking anomaly detection methods. To address this gap, we introduce a new high-dimensional dataset from IBM Cloud, collected over 4.5 months from the IBM Cloud Console. This dataset comprises 39,365 rows and 117,448 columns of telemetry data. Additionally, we demonstrate the application of machine learning models for anomaly detection and discuss the key challenges faced in this process. This study and the accompanying dataset provide a resource for researchers and practitioners in cloud system monitoring. It facilitates more efficient testing of anomaly detection methods in real-world data, helping to advance the development of robust solutions to maintain the health and performance of large-scale cloud infrastructures.
Authors: Mahammad Humayoo
Abstract: In numerous episodic reinforcement learning (RL) settings, SARSA-based methodologies are employed to enhance policies aimed at maximizing returns over long horizons. Conventional SARSA algorithms, however, have difficulties in balancing bias and variation due to the reliance on a singular, fixed discount factor. This study expands the temporal difference decomposition approach, TD($\Delta$), to the SARSA algorithm, which we designate as SARSA($\Delta$). SARSA, a widely utilised on-policy RL method, enhances action-value functions via temporal difference updates. TD($\Delta$) facilitates learning over several time-scales by breaking the action-value function into components associated with distinct discount factors. This decomposition improves learning efficiency and stability, particularly in problems necessitating long-horizon optimization. We illustrate that our methodology mitigates bias in SARSA's updates while facilitating accelerated convergence in both deterministic and stochastic environments. Experimental findings across many benchmark tasks indicate that the proposed SARSA($\Delta$) surpasses conventional TD learning methods in both tabular and deep RL environments.
Authors: Mohammad Hassan Heydari, Arshia Hemmat, Erfan Naman, Afsaneh Fatemi
Abstract: Retrieval Augmented Generation (RAG) has emerged as a widely adopted approach to mitigate the limitations of large language models (LLMs) in answering domain-specific questions. Previous research has predominantly focused on improving the accuracy and quality of retrieved data chunks to enhance the overall performance of the generation pipeline. However, despite ongoing advancements, the critical issue of retrieving irrelevant information -- which can impair the ability of the model to utilize its internal knowledge effectively -- has received minimal attention. In this work, we investigate the impact of retrieving irrelevant information in open-domain question answering, highlighting its significant detrimental effect on the quality of LLM outputs. To address this challenge, we propose the Context Awareness Gate (CAG) architecture, a novel mechanism that dynamically adjusts the LLMs' input prompt based on whether the user query necessitates external context retrieval. Additionally, we introduce the Vector Candidates method, a core mathematical component of CAG that is statistical, LLM-independent, and highly scalable. We further examine the distributions of relationships between contexts and questions, presenting a statistical analysis of these distributions. This analysis can be leveraged to enhance the context retrieval process in Retrieval Augmented Generation (RAG) systems.
Authors: Minzhe Tan, Xinlin Fan, Jian He, Yi Hou, Zhan Liu, Yaopeng Jiang, Y. M. Jiang
Abstract: This paper introduces a high-performance artificial intelligence operating system tailored for low-altitude aviation, designed to address key challenges such as real-time task execution, computational efficiency, and seamless modular collaboration. Built on a powerful hardware platform and leveraging the UNIX architecture, the system implements a distributed data processing strategy that ensures rapid and efficient synchronization across critical modules, including vision, navigation, and perception. By adopting dynamic resource management, it optimally allocates computational resources, such as CPU and GPU, based on task priority and workload, ensuring high performance for demanding tasks like real-time video processing and AI model inference. Furthermore, the system features an advanced interrupt handling mechanism that allows for quick responses to sudden environmental changes, such as obstacle detection, by prioritizing critical tasks, thus improving safety and mission success rates. Robust security measures, including data encryption, access control, and fault tolerance, ensure the system's resilience against external threats and its ability to recover from potential hardware or software failures. Complementing these core features are modular components for image analysis, multi-sensor fusion, dynamic path planning, multi-drone coordination, and ground station monitoring. Additionally, a low-code development platform simplifies user customization, making the system adaptable to various mission-specific needs. This comprehensive approach ensures the system meets the evolving demands of intelligent aviation, providing a stable, efficient, and secure environment for complex drone operations.
Authors: Dongeon Lee, Sunwoong Yang, Jae-Won Oh, Su-Gil Cho, Sanghyuk Kim, Namwoo Kang
Abstract: Environmental pollution and fossil fuel depletion have prompted the need for renewable energy-based power generation. However, its stability is often challenged by low energy density and non-stationary conditions. Wave energy converters (WECs), in particular, need reliable real-time wave height prediction to address these issues caused by irregular wave patterns, which can lead to the inefficient and unstable operation of WECs. In this study, we propose an AI-powered reliable real-time wave height prediction model that integrates long short-term memory (LSTM) networks for temporal prediction with deep ensemble (DE) for robust uncertainty quantification (UQ), ensuring high accuracy and reliability. To further enhance the reliability, uncertainty calibration is applied, which has proven to significantly improve the quality of the quantified uncertainty. Using real operational data from an oscillating water column-wave energy converter (OWC-WEC) system in Jeju, South Korea, the model achieves notable accuracy (R2 > 0.9), while increasing uncertainty quality by over 50% through simple calibration technique. Furthermore, a comprehensive parametric study is conducted to explore the effects of key model hyperparameters, offering valuable guidelines for diverse operational scenarios, characterized by differences in wavelength, amplitude, and period. These results demonstrate the model's capability to deliver reliable predictions, facilitating digital twin of the ocean.
Authors: Yuhang He, Yash Jain, Xubo Liu, Andrew Markham, Vibhav Vineet
Abstract: Despite significant advancements in Text-to-Audio (TTA) generation models achieving high-fidelity audio with fine-grained context understanding, they struggle to model the relations between audio events described in the input text. However, previous TTA methods have not systematically explored audio event relation modeling, nor have they proposed frameworks to enhance this capability. In this work, we systematically study audio event relation modeling in TTA generation models. We first establish a benchmark for this task by: 1. proposing a comprehensive relation corpus covering all potential relations in real-world scenarios; 2. introducing a new audio event corpus encompassing commonly heard audios; and 3. proposing new evaluation metrics to assess audio event relation modeling from various perspectives. Furthermore, we propose a finetuning framework to enhance existing TTA models ability to model audio events relation. Code is available at: https://github.com/yuhanghe01/RiTTA
Authors: Jinhyeok Choi, Heehyeon Kim, Joyce Jiyoung Whang
Abstract: Graph neural networks (GNNs) have emerged as an effective tool for fraud detection, identifying fraudulent users, and uncovering malicious behaviors. However, attacks against GNN-based fraud detectors and their risks have rarely been studied, thereby leaving potential threats unaddressed. Recent findings suggest that frauds are increasingly organized as gangs or groups. In this work, we design attack scenarios where fraud gangs aim to make their fraud nodes misclassified as benign by camouflaging their illicit activities in collusion. Based on these scenarios, we study adversarial attacks against GNN-based fraud detectors by simulating attacks of fraud gangs in three real-world fraud cases: spam reviews, fake news, and medical insurance frauds. We define these attacks as multi-target graph injection attacks and propose MonTi, a transformer-based Multi-target one-Time graph injection attack model. MonTi simultaneously generates attributes and edges of all attack nodes with a transformer encoder, capturing interdependencies between attributes and edges more effectively than most existing graph injection attack methods that generate these elements sequentially. Additionally, MonTi adaptively allocates the degree budget for each attack node to explore diverse injection structures involving target, candidate, and attack nodes, unlike existing methods that fix the degree budget across all attack nodes. Experiments show that MonTi outperforms the state-of-the-art graph injection attack methods on five real-world graphs.
Authors: Xinyu Yang, Yu Sun, Xinyang Chen, Ying Zhang, Xiaojie Yuan
Abstract: Spatial-temporal data collected across different geographic locations often suffer from missing values, posing challenges to data analysis. Existing methods primarily leverage fixed spatial graphs to impute missing values, which implicitly assume that the spatial relationship is roughly the same for all features across different locations. However, they may overlook the different spatial relationships of diverse features recorded by sensors in different locations. To address this, we introduce the multi-scale Graph Structure Learning framework for spatial-temporal Imputation (GSLI) that dynamically adapts to the heterogeneous spatial correlations. Our framework encompasses node-scale graph structure learning to cater to the distinct global spatial correlations of different features, and feature-scale graph structure learning to unveil common spatial correlation across features within all stations. Integrated with prominence modeling, our framework emphasizes nodes and features with greater significance in the imputation process. Furthermore, GSLI incorporates cross-feature and cross-temporal representation learning to capture spatial-temporal dependencies. Evaluated on six real incomplete spatial-temporal datasets, GSLI showcases the improvement in data imputation.
Authors: Jia-Hong Huang, Yixian Shen, Hongyi Zhu, Stevan Rudinac, Evangelos Kanoulas
Abstract: Large Language Models (LLMs) have shown remarkable performance across various tasks, but the escalating demands on computational resources pose significant challenges, particularly in the extensive utilization of full fine-tuning for downstream tasks. To address this, parameter-efficient fine-tuning (PEFT) methods have been developed, but they often underperform compared to full fine-tuning and struggle with memory efficiency. In this work, we introduce Gradient Weight-Normalized Low-Rank Projection (GradNormLoRP), a novel approach that enhances both parameter and memory efficiency while maintaining comparable performance to full fine-tuning. GradNormLoRP normalizes the weight matrix to improve gradient conditioning, facilitating better convergence during optimization. Additionally, it applies low-rank approximations to the weight and gradient matrices, significantly reducing memory usage during training. Extensive experiments demonstrate that our 8-bit GradNormLoRP reduces optimizer memory usage by up to 89.5% and enables the pre-training of large LLMs, such as LLaMA 7B, on consumer-level GPUs like the NVIDIA RTX 4090, without additional inference costs. Moreover, GradNormLoRP outperforms existing low-rank methods in fine-tuning tasks. For instance, when fine-tuning the RoBERTa model on all GLUE tasks with a rank of 8, GradNormLoRP achieves an average score of 80.65, surpassing LoRA's score of 79.23. These results underscore GradNormLoRP as a promising alternative for efficient LLM pre-training and fine-tuning. Source code: https://github.com/Jhhuangkay/Gradient-Weight-normalized-Low-rank-Projection-for-Efficient-LLM-Training
Authors: En Fu, Yanyan Hu
Abstract: Contrastive learning underpins most current self-supervised time series representation methods. The strategy for constructing positive and negative sample pairs significantly affects the final representation quality. However, due to the continuous nature of time series semantics, the modeling approach of contrastive learning struggles to accommodate the characteristics of time series data. This results in issues such as difficulties in constructing hard negative samples and the potential introduction of inappropriate biases during positive sample construction. Although some recent works have developed several scientific strategies for constructing positive and negative sample pairs with improved effectiveness, they remain constrained by the contrastive learning framework. To fundamentally overcome the limitations of contrastive learning, this paper introduces Frequency-masked Embedding Inference (FEI), a novel non-contrastive method that completely eliminates the need for positive and negative samples. The proposed FEI constructs 2 inference branches based on a prompting strategy: 1) Using frequency masking as prompts to infer the embedding representation of the target series with missing frequency bands in the embedding space, and 2) Using the target series as prompts to infer its frequency masking embedding. In this way, FEI enables continuous semantic relationship modeling for time series. Experiments on 8 widely used time series datasets for classification and regression tasks, using linear evaluation and end-to-end fine-tuning, show that FEI significantly outperforms existing contrastive-based methods in terms of generalization. This study provides new insights into self-supervised representation learning for time series. The code is available at https://github.com/USTBInnovationPark/Frequency-masked-Embedding-Inference.
URLs: https://github.com/USTBInnovationPark/Frequency-masked-Embedding-Inference.
Authors: Yuan Mi, Pu Ren, Hongteng Xu, Hongsheng Liu, Zidong Wang, Yike Guo, Ji-Rong Wen, Hao Sun, Yang Liu
Abstract: Data-centric methods have shown great potential in understanding and predicting spatiotemporal dynamics, enabling better design and control of the object system. However, deep learning models often lack interpretability, fail to obey intrinsic physics, and struggle to cope with the various domains. While geometry-based methods, e.g., graph neural networks (GNNs), have been proposed to further tackle these challenges, they still need to find the implicit physical laws from large datasets and rely excessively on rich labeled data. In this paper, we herein introduce the conservation-informed GNN (CiGNN), an end-to-end explainable learning framework, to learn spatiotemporal dynamics based on limited training data. The network is designed to conform to the general conservation law via symmetry, where conservative and non-conservative information passes over a multiscale space enhanced by a latent temporal marching strategy. The efficacy of our model has been verified in various spatiotemporal systems based on synthetic and real-world datasets, showing superiority over baseline models. Results demonstrate that CiGNN exhibits remarkable accuracy and generalizability, and is readily applicable to learning for prediction of various spatiotemporal dynamics in a spatial domain with complex geometry.
Authors: Ziang Chen, Zhengjiang Lin, Shi Chen, Yury Polyanskiy, Philippe Rigollet
Abstract: Graph neural networks (GNNs) have achieved remarkable empirical success in processing and representing graph-structured data across various domains. However, a significant challenge known as "oversmoothing" persists, where vertex features become nearly indistinguishable in deep GNNs, severely restricting their expressive power and practical utility. In this work, we analyze the asymptotic oversmoothing rates of deep GNNs with and without residual connections by deriving explicit convergence rates for a normalized vertex similarity measure. Our analytical framework is grounded in the multiplicative ergodic theorem. Furthermore, we demonstrate that adding residual connections effectively mitigates or prevents oversmoothing across several broad families of parameter distributions. The theoretical findings are strongly supported by numerical experiments.
Authors: Zhenyu Guo, Wenguang Chen
Abstract: Transformers have achieved remarkable success across diverse domains, but their monolithic architecture presents challenges in interpretability, adaptability, and scalability. This paper introduces a novel modular Transformer architecture that explicitly decouples knowledge and reasoning through a generalized cross-attention mechanism to a globally shared knowledge base with layer-specific transformations, specifically designed for effective knowledge retrieval. Critically, we provide a rigorous mathematical derivation demonstrating that the Feed-Forward Network (FFN) in a standard Transformer is a specialized case (a closure) of this generalized cross-attention, revealing its role in implicit knowledge retrieval and validating our design. This theoretical framework provides a new lens for understanding FFNs and lays the foundation for future research exploring enhanced interpretability, adaptability, and scalability, enabling richer interplay with external knowledge bases and other systems.
Authors: Xiaoyi Wu, Bo Ji, Bin Li
Abstract: Combinatorial Multi-Armed Bandit with fairness constraints is a framework where multiple arms form a super arm and can be pulled in each round under uncertainty to maximize cumulative rewards while ensuring the minimum average reward required by each arm. The existing pessimistic-optimistic algorithm linearly combines virtual queue-lengths (tracking the fairness violations) and Upper Confidence Bound estimates as a weight for each arm and selects a super arm with the maximum total weight. The number of super arms could be exponential to the number of arms in many scenarios. In wireless networks, interference constraints can cause the number of super arms to grow exponentially with the number of arms. Evaluating all the feasible super arms to find the one with the maximum total weight can incur extremely high computational complexity in the pessimistic-optimistic algorithm. To avoid this, we develop a low-complexity fair learning algorithm based on the so-called pick-and-compare approach that involves randomly picking $M$ feasible super arms to evaluate. By setting $M$ to a constant, the number of comparison steps in the pessimistic-optimistic algorithm can be reduced to a constant, thereby significantly reducing the computational complexity. Our theoretical proof shows this low-complexity design incurs only a slight sacrifice in fairness and regret performance. Finally, we validate the theoretical result by extensive simulations.
Authors: Roberto Dias Algarte
Abstract: This article presents a generic approach to convolution that significantly differs from conventional methodologies in the current Machine Learning literature. The approach, in its mathematical aspects, proved to be clear and concise, particularly when high-order tensors are involved. In this context, a rational theory of regression in neural networks is developed, as a framework for a generic view of sparse convolutional neural networks, the primary focus of this study. As a direct outcome, the classic Backpropagation Algorithm is redefined to align with this rational tensor-based approach and presented in its simplest, most generic form.
Authors: Cheng Wan, Runkai Tao, Zheng Du, Yang Katie Zhao, Yingyan Celine Lin
Abstract: Graph convolutional networks (GCNs) have demonstrated superiority in graph-based learning tasks. However, training GCNs on full graphs is particularly challenging, due to the following two challenges: (1) the associated feature tensors can easily explode the memory and block the communication bandwidth of modern accelerators, and (2) the computation workflow in training GCNs alternates between sparse and dense matrix operations, complicating the efficient utilization of computational resources. Existing solutions for scalable distributed full-graph GCN training mostly adopt partition parallelism, which is unsatisfactory as they only partially address the first challenge while incurring scaled-out communication volume. To this end, we propose MixGCN aiming to simultaneously address both the aforementioned challenges towards GCN training. To tackle the first challenge, MixGCN integrates mixture of parallelism. Both theoretical and empirical analysis verify its constant communication volumes and enhanced balanced workload; For handling the second challenge, we consider mixture of accelerators (i.e., sparse and dense accelerators) with a dedicated accelerator for GCN training and a fine-grain pipeline. Extensive experiments show that MixGCN achieves boosted training efficiency and scalability.
Authors: Yonggan Fu, Wuyang Chen, Haotao Wang, Haoran Li, Yingyan Celine Lin, Zhangyang Wang
Abstract: The compression of Generative Adversarial Networks (GANs) has lately drawn attention, due to the increasing demand for deploying GANs into mobile devices for numerous applications such as image translation, enhancement and editing. However, compared to the substantial efforts to compressing other deep models, the research on compressing GANs (usually the generators) remains at its infancy stage. Existing GAN compression algorithms are limited to handling specific GAN architectures and losses. Inspired by the recent success of AutoML in deep compression, we introduce AutoML to GAN compression and develop an AutoGAN-Distiller (AGD) framework. Starting with a specifically designed efficient search space, AGD performs an end-to-end discovery for new efficient generators, given the target computational resource constraints. The search is guided by the original GAN model via knowledge distillation, therefore fulfilling the compression. AGD is fully automatic, standalone (i.e., needing no trained discriminators), and generically applicable to various GAN models. We evaluate AGD in two representative GAN tasks: image translation and super resolution. Without bells and whistles, AGD yields remarkably lightweight yet more competitive compressed models, that largely outperform existing alternatives. Our codes and pretrained models are available at https://github.com/TAMU-VITA/AGD.
Authors: Marcus N. Boon, Lorenzo Cassola, Hans-Christian Ruiz Euler, Tao Chen, Bram van de Ven, Unai Alegre Ibarra, Peter A. Bobbert, Wilfred G. van der Wiel
Abstract: Deep learning, a multi-layered neural network approach inspired by the brain, has revolutionized machine learning. One of its key enablers has been backpropagation, an algorithm that computes the gradient of a loss function with respect to the weights and biases in the neural network model, in combination with its use in gradient descent. However, the implementation of deep learning in digital computers is intrinsically energy hungry, with energy consumption becoming prohibitively high for many applications. This has stimulated the development of specialized hardware, ranging from neuromorphic CMOS integrated circuits and integrated photonic tensor cores to unconventional, material-based computing system. The learning process in these material systems, realized, e.g., by artificial evolution, equilibrium propagation or surrogate modelling, is a complicated and time-consuming process. Here, we demonstrate a simple yet efficient and accurate gradient extraction method, based on the principle of homodyne detection, for performing gradient descent on a loss function directly in a physical system without the need of an analytical description. By perturbing the parameters that need to be optimized using sinusoidal waveforms with distinct frequencies, we effectively obtain the gradient information in a highly robust and scalable manner. We illustrate the method in dopant network processing units, but argue that it is applicable in a wide range of physical systems. Homodyne gradient extraction can in principle be fully implemented in materia, facilitating the development of autonomously learning material systems.
Authors: Mengbing Li, Chengchun Shi, Zhenke Wu, Piotr Fryzlewicz
Abstract: We consider offline reinforcement learning (RL) methods in possibly nonstationary environments. Many existing RL algorithms in the literature rely on the stationarity assumption that requires the system transition and the reward function to be constant over time. However, the stationarity assumption is restrictive in practice and is likely to be violated in a number of applications, including traffic signal control, robotics and mobile health. In this paper, we develop a consistent procedure to test the nonstationarity of the optimal Q-function based on pre-collected historical data, without additional online data collection. Based on the proposed test, we further develop a sequential change point detection method that can be naturally coupled with existing state-of-the-art RL methods for policy optimization in nonstationary environments. The usefulness of our method is illustrated by theoretical results, simulation studies, and a real data example from the 2018 Intern Health Study. A Python implementation of the proposed procedure is available at https://github.com/limengbinggz/CUSUM-RL.
Authors: Shijun Liang, Anish Lahiri, Saiprasad Ravishankar
Abstract: Recent medical image reconstruction techniques focus on generating high-quality medical images suitable for clinical use at the lowest possible cost and with the fewest possible adverse effects on patients. Recent works have shown significant promise for reconstructing MR images from sparsely sampled k-space data using deep learning. In this work, we propose a technique that rapidly estimates deep neural networks directly at reconstruction time by fitting them on small adaptively estimated neighborhoods of a training set. In brief, our algorithm alternates between searching for neighbors in a data set that are similar to the test reconstruction, and training a local network on these neighbors followed by updating the test reconstruction. Because our reconstruction model is learned on a dataset that is in some sense similar to the image being reconstructed rather than being fit on a large, diverse training set, it is more adaptive to new scans. It can also handle changes in training sets and flexible scan settings, while being relatively fast. Our approach, dubbed LONDN-MRI, was validated on multiple data sets using deep unrolled reconstruction networks. Reconstructions were performed at four fold and eight fold undersampling of k-space with 1D variable-density random phase-encode undersampling masks. Our results demonstrate that our proposed locally-trained method produces higher-quality reconstructions compared to models trained globally on larger datasets as well as other scan-adaptive methods.
Authors: Krist\'of N\'emeth, D\'aniel Hadh\'azi
Abstract: We apply artificial neural networks (ANNs) to nowcast quarterly GDP growth for the U.S. economy. Using the monthly FRED-MD database, we compare the nowcasting performance of five different ANN architectures: the multilayer perceptron (MLP), the one-dimensional convolutional neural network (1D CNN), the Elman recurrent neural network (RNN), the long short-term memory network (LSTM), and the gated recurrent unit (GRU). The empirical analysis presents results from two distinctively different evaluation periods. The first (2012:Q1 -- 2019:Q4) is characterized by balanced economic growth, while the second (2012:Q1 -- 2024:Q2) also includes periods of the COVID-19 recession. During the first evaluation period, longer input sequences slightly improve nowcasting performance for some ANNs, but the best accuracy is still achieved with 8-month-long input sequences at the end of the nowcasting window. Results from the second test period depict the role of long-term memory even more clearly. The MLP, the 1D CNN, and the Elman RNN work best with 8-month-long input sequences at each step of the nowcasting window. The relatively weak performance of the gated RNNs also suggests that architectural features enabling long-term memory do not result in more accurate nowcasts for GDP growth. The combined results indicate that the 1D CNN seems to represent a \textit{``sweet spot''} between the simple time-agnostic MLP and the more complex (gated) RNNs. The network generates nearly as accurate nowcasts as the best competitor for the first test period, while it achieves the overall best accuracy during the second evaluation period. Consequently, as a first in the literature, we propose the application of the 1D CNN for economic nowcasting.
Authors: Renhao Wang, Yu Sun, Arnuv Tandon, Yossi Gandelsman, Xinlei Chen, Alexei A. Efros, Xiaolong Wang
Abstract: Prior work has established Test-Time Training (TTT) as a general framework to further improve a trained model at test time. Before making a prediction on each test instance, the model is first trained on the same instance using a self-supervised task such as reconstruction. We extend TTT to the streaming setting, where multiple test instances - video frames in our case - arrive in temporal order. Our extension is online TTT: The current model is initialized from the previous model, then trained on the current frame and a small window of frames immediately before. Online TTT significantly outperforms the fixed-model baseline for four tasks, on three real-world datasets. The improvements are more than 2.2x and 1.5x for instance and panoptic segmentation. Surprisingly, online TTT also outperforms its offline variant that accesses strictly more information, training on all frames from the entire test video regardless of temporal order. This finding challenges those in prior work using synthetic videos. We formalize a notion of locality as the advantage of online over offline TTT, and analyze its role with ablations and a theory based on bias-variance trade-off.
Authors: Benjamin Steenhoek, Michele Tufano, Neel Sundaresan, Alexey Svyatkovskiy
Abstract: Software testing is a crucial aspect of software development, and the creation of high-quality tests that adhere to best practices is essential for effective maintenance. Recently, Large Language Models (LLMs) have gained popularity for code generation, including the automated creation of test cases. However, these LLMs are often trained on vast amounts of publicly available code, which may include test cases that do not adhere to best practices and may even contain test smells (anti-patterns). To address this issue, we propose a novel technique called Reinforcement Learning from Static Quality Metrics (RLSQM). To begin, we analyze the anti-patterns generated by the LLM and show that LLMs can generate undesirable test smells. Thus, we train specific reward models for each static quality metric, then utilize Proximal Policy Optimization (PPO) to train models for optimizing a single quality metric at a time. Furthermore, we amalgamate these rewards into a unified reward model aimed at capturing different best practices and quality aspects of tests. By comparing RL-trained models with those trained using supervised learning, we provide insights into how reliably utilize RL to improve test generation quality and into the effects of various training strategies. Our experimental results demonstrate that the RL-optimized model consistently generated high-quality test cases compared to the base LLM, improving the model by up to 21%, and successfully generates nearly 100% syntactically correct code. RLSQM also outperformed GPT-4 on four out of seven metrics. This represents a significant step towards enhancing the overall efficiency and reliability of software testing through Reinforcement Learning and static quality metrics. Our data are available at https://figshare.com/s/ded476c8d4c221222849.
Authors: Zonghai Yao, Benjamin J Schloss, Sai P. Selvaraj
Abstract: Recent work has shown the promise of learning with human feedback paradigms to produce human-determined high-quality text. Existing works use human feedback to train large language models (LLMs) in general domain abstractive summarization and have obtained summary quality exceeding traditional likelihood training. In this paper, we focus on a less explored form of human feedback -- Human Edits. We propose Sequence Alignment (un)Likelihood Training (SALT), a novel technique to use both the human-edited and model-generated data together in the training loop. In addition, we demonstrate simulating Human Edits with ground truth summaries coming from existing training data -- Imitation edits, along with the model-generated summaries obtained after the training, to reduce the need for expensive human-edit data. In our experiments, we extend human feedback exploration from general domain summarization to medical domain summarization. Our results demonstrate the effectiveness of SALT in improving the summary quality with Human and Imitation Edits. Through additional experiments, we show that SALT outperforms the conventional RLHF method (designed for human preferences) -- DPO, when applied to human-edit data. We hope the evidence in our paper prompts researchers to explore, collect, and better use different human feedback approaches scalably.
Authors: Shohei Wakayama, Nisar Ahmed
Abstract: We introduce a new variant of contextual multi-armed bandits (CMABs) called observation-augmented CMABs (OA-CMABs) wherein a robot uses extra outcome observations from an external information source, e.g. humans. In OA-CMABs, external observations are a function of context features and thus provide evidence on top of observed option outcomes to infer hidden parameters. However, if external data is error-prone, measures must be taken to preserve the correctness of inference. To this end, we derive a robust Bayesian inference process for OA-CMABs based on recently developed probabilistic semantic data association techniques, which handle complex mixture model parameter priors and hybrid discrete-continuous observation likelihoods for semantic external data sources. To cope with combined uncertainties in OA-CMABs, we also derive a new active inference algorithm for optimal option selection based on approximate expected free energy minimization. This generalizes prior work on CMAB active inference by accounting for faulty observations and non-Gaussian distributions. Results for a simulated deep space search site selection problem show that, even if incorrect semantic observations are provided externally, e.g. by scientists, efficient decision-making and robust parameter inference are still achieved in a wide variety of conditions.
Authors: Weijia Zhang, Jindong Han, Zhao Xu, Hang Ni, Tengfei Lyu, Hao Liu, Hui Xiong
Abstract: The integration of machine learning techniques has become a cornerstone in the development of intelligent urban services, significantly contributing to the enhancement of urban efficiency, sustainability, and overall livability. Recent advancements in foundational models, such as ChatGPT, have introduced a paradigm shift within the fields of machine learning and artificial intelligence. These models, with their exceptional capacity for contextual comprehension, problem-solving, and task adaptability, present a transformative opportunity to reshape the future of smart cities and drive progress toward Urban General Intelligence (UGI). Despite increasing attention to Urban Foundation Models (UFMs), this rapidly evolving field faces critical challenges, including the lack of clear definitions, systematic reviews, and universalizable solutions. To address these issues, this paper first introduces the definition and concept of UFMs and highlights the distinctive challenges involved in their development. Furthermore, we present a data-centric taxonomy that classifies existing research on UFMs according to the various urban data modalities and types. In addition, we propose a prospective framework designed to facilitate the realization of versatile UFMs, aimed at overcoming the identified challenges and driving further progress in this field. Finally, this paper explores the wide-ranging applications of UFMs within urban contexts, illustrating their potential to significantly impact and transform urban systems. A comprehensive collection of relevant research papers and open-source resources have been collated and are continuously updated at: https://github.com/usail-hkust/Awesome-Urban-Foundation-Models.
URLs: https://github.com/usail-hkust/Awesome-Urban-Foundation-Models.
Authors: Zolnamar Dorjsembe, Hsing-Kuo Pao, Furen Xiao
Abstract: This study introduces Polyp-DDPM, a diffusion-based method for generating realistic images of polyps conditioned on masks, aimed at enhancing the segmentation of gastrointestinal (GI) tract polyps. Our approach addresses the challenges of data limitations, high annotation costs, and privacy concerns associated with medical images. By conditioning the diffusion model on segmentation masks-binary masks that represent abnormal areas-Polyp-DDPM outperforms state-of-the-art methods in terms of image quality (achieving a Frechet Inception Distance (FID) score of 78.47, compared to scores above 83.79) and segmentation performance (achieving an Intersection over Union (IoU) of 0.7156, versus less than 0.6694 for synthetic images from baseline models and 0.7067 for real data). Our method generates a high-quality, diverse synthetic dataset for training, thereby enhancing polyp segmentation models to be comparable with real images and offering greater data augmentation capabilities to improve segmentation models. The source code and pretrained weights for Polyp-DDPM are made publicly available at https://github.com/mobaidoctor/polyp-ddpm.
Authors: Kim Hammar, Tao Li, Rolf Stadler, Quanyan Zhu
Abstract: We study automated security response for an IT infrastructure and formulate the interaction between an attacker and a defender as a partially observed, non-stationary game. We relax the standard assumption that the game model is correctly specified and consider that each player has a probabilistic conjecture about the model, which may be misspecified in the sense that the true model has probability 0. This formulation allows us to capture uncertainty and misconception about the infrastructure and the intents of the players. To learn effective game strategies online, we design Conjectural Online Learning (COL), a novel method where a player iteratively adapts its conjecture using Bayesian learning and updates its strategy through rollout. We prove that the conjectures converge to best fits, and we provide a bound on the performance improvement that rollout enables with a conjectured model. To characterize the steady state of the game, we propose a variant of the Berk-Nash equilibrium. We present COL through an advanced persistent threat use case. Testbed evaluations show that COL produces effective security strategies that adapt to a changing environment. We also find that COL enables faster convergence than current reinforcement learning techniques.
Authors: Paul Dommel
Abstract: Kernel ridge regression, in general, is expensive in memory allocation and computation time. This paper addresses low rank approximations and surrogates for kernel ridge regression, which bridge these difficulties. The fundamental contribution of the paper is a lower bound on the minimal rank such that the prediction power of the approximation remains reliable. Based on this bound, we demonstrate that the computational cost of the most popular low rank approach, which is the Nystr\"om method, is almost linear in the sample size. This justifies the method from a theoretical point of view. Moreover, the paper provides a significant extension of the feasible choices of the regularization parameter. The result builds on a thorough theoretical analysis of the approximation of elementary kernel functions by elements in the range of the associated integral operator. We provide estimates of the approximation error and characterize the behavior of the norm of the underlying weight function.
Authors: B. N. Kausik
Abstract: Trained LLMs are typically sparse in that most of the parameters are zero, raising questions on efficiency. In response, we inquire into efficient LLMs, i.e. those with the fewest parameters that achieve the desired accuracy on a training corpus. Specifically, we compare theoretical and empirical estimates for training loss to obtain upper and lower bounds on the number of unique sequences in a natural training corpus as a function of its size. Our result implies (1) to double the number of skills represented in a training corpus, the corpus must scale roughly eighteen fold (2) for efficient LLMs, the number of parameters N and the size D of a natural training corpus scale as $N \propto D^{0.24} (3) if the number of parameters of an LLM is smaller than the number of unique sequences in the training corpus, scaling up can uncover emergent skills.
Authors: Kechun Xu, Zhongxiang Zhou, Jun Wu, Haojian Lu, Rong Xiong, Yue Wang
Abstract: We focus on the task of unknown object rearrangement, where a robot is supposed to re-configure the objects into a desired goal configuration specified by an RGB-D image. Recent works explore unknown object rearrangement systems by incorporating learning-based perception modules. However, they are sensitive to perception error, and pay less attention to task-level performance. In this paper, we aim to develop an effective system for unknown object rearrangement amidst perception noise. We theoretically reveal that the noisy perception impacts grasp and place in a decoupled way, and show such a decoupled structure is valuable to improve task optimality. We propose GSP, a dual-loop system with the decoupled structure as prior. For the inner loop, we learn a see policy for self-confident in-hand object matching. For the outer loop, we learn a grasp policy aware of object matching and grasp capability guided by task-level rewards. We leverage the foundation model CLIP for object matching, policy learning and self-termination. A series of experiments indicate that GSP can conduct unknown object rearrangement with higher completion rates and fewer steps.
Authors: Yongqi Wang, Ruofan Hu, Rongjie Huang, Zhiqing Hong, Ruiqi Li, Wenrui Liu, Fuming You, Tao Jin, Zhou Zhao
Abstract: Recent singing-voice-synthesis (SVS) methods have achieved remarkable audio quality and naturalness, yet they lack the capability to control the style attributes of the synthesized singing explicitly. We propose Prompt-Singer, the first SVS method that enables attribute controlling on singer gender, vocal range and volume with natural language. We adopt a model architecture based on a decoder-only transformer with a multi-scale hierarchy, and design a range-melody decoupled pitch representation that enables text-conditioned vocal range control while keeping melodic accuracy. Furthermore, we explore various experiment settings, including different types of text representations, text encoder fine-tuning, and introducing speech data to alleviate data scarcity, aiming to facilitate further research. Experiments show that our model achieves favorable controlling ability and audio quality. Audio samples are available at http://prompt-singer.github.io .
Authors: Salwa Mostafa, Mateus P. Mota, Alvaro Valcarce, Mehdi Bennis
Abstract: We investigate the problem of supporting Industrial Internet of Things user equipment (IIoT UEs) with intent (i.e., requested quality of service (QoS)) and random traffic arrival. A deep reinforcement learning (DRL) based centralized dynamic scheduler for time-frequency resources is proposed to learn how to schedule the available communication resources among the IIoT UEs. The proposed scheduler leverages an RL framework to adapt to the dynamic changes in the wireless communication system and traffic arrivals. Moreover, a graph-based reduction scheme is proposed to reduce the state and action space of the RL framework to allow fast convergence and a better learning strategy. Simulation results demonstrate the effectiveness of the proposed intelligent scheduler in guaranteeing the expressed intent of IIoT UEs compared to several traditional scheduling schemes, such as round-robin, semi-static, and heuristic approaches. The proposed scheduler also outperforms the contention-free and contention-based schemes in maximizing the number of successfully computed tasks.
Authors: Hong Ye Tan, Ziruo Cai, Marcelo Pereyra, Subhadip Mukherjee, Junqi Tang, Carola-Bibiane Sch\"onlieb
Abstract: Imaging is a standard example of an inverse problem, where the task of reconstructing a ground truth from a noisy measurement is ill-posed. Recent state-of-the-art approaches for imaging use deep learning, spearheaded by unrolled and end-to-end models and trained on various image datasets. However, many such methods require the availability of ground truth data, which may be unavailable or expensive, leading to a fundamental barrier that can not be bypassed by choice of architecture. Unsupervised learning presents an alternative paradigm that bypasses this requirement, as they can be learned directly on noisy data and do not require any ground truths. A principled Bayesian approach to unsupervised learning is to maximize the marginal likelihood with respect to the given noisy measurements, which is intrinsically linked to classical variational regularization. We propose an unsupervised approach using maximum marginal likelihood estimation to train a convex neural network-based image regularization term directly on noisy measurements, improving upon previous work in both model expressiveness and dataset size. Experiments demonstrate that the proposed method produces priors that are near competitive when compared to the analogous supervised training method for various image corruption operators, maintaining significantly better generalization properties when compared to end-to-end methods. Moreover, we provide a detailed theoretical analysis of the convergence properties of our proposed algorithm.
Authors: Wenyi Xiao, Ziwei Huang, Leilei Gan, Wanggui He, Haoyuan Li, Zhelun Yu, Fangxun Shu, Hao Jiang, Linchao Zhu
Abstract: The rapidly developing Large Vision Language Models (LVLMs) have shown notable capabilities on a range of multi-modal tasks, but still face the hallucination phenomena where the generated texts do not align with the given contexts, significantly restricting the usages of LVLMs. Most previous work detects and mitigates hallucination at the coarse-grained level or requires expensive annotation (e.g., labeling by proprietary models or human experts). To address these issues, we propose detecting and mitigating hallucinations in LVLMs via fine-grained AI feedback. The basic idea is that we generate a small-size sentence-level hallucination annotation dataset by proprietary models, whereby we train a hallucination detection model which can perform sentence-level hallucination detection, covering primary hallucination types (i.e., object, attribute, and relationship). Then, we propose a detect-then-rewrite pipeline to automatically construct preference dataset for training hallucination mitigating model. Furthermore, we propose differentiating the severity of hallucinations, and introducing a Hallucination Severity-Aware Direct Preference Optimization (HSA-DPO) for mitigating hallucination in LVLMs by incorporating the severity of hallucinations into preference learning. Extensive experiments demonstrate the effectiveness of our method.
Authors: Ariel Neufeld, Philipp Schmocker, Sizhou Wu
Abstract: In this paper, we present a randomized extension of the deep splitting algorithm introduced in [Beck, Becker, Cheridito, Jentzen, and Neufeld (2021)] using random neural networks suitable to approximately solve both high-dimensional nonlinear parabolic PDEs and PIDEs with jumps having (possibly) infinite activity. We provide a full error analysis of our so-called random deep splitting method. In particular, we prove that our random deep splitting method converges to the (unique viscosity) solution of the nonlinear PDE or PIDE under consideration. Moreover, we empirically analyze our random deep splitting method by considering several numerical examples including both nonlinear PDEs and nonlinear PIDEs relevant in the context of pricing of financial derivatives under default risk. In particular, we empirically demonstrate in all examples that our random deep splitting method can approximately solve nonlinear PDEs and PIDEs in 10'000 dimensions within seconds.
Authors: Chenyu Huang, Zhengyang Tang, Shixi Hu, Ruoqing Jiang, Xin Zheng, Dongdong Ge, Benyou Wang, Zizhuo Wang
Abstract: Optimization modeling plays a critical role in the application of Operations Research (OR) tools to address real-world problems, yet they pose challenges and require extensive expertise from OR experts. With the advent of large language models (LLMs), new opportunities have emerged to streamline and automate such task. However, current research predominantly relies on closed-source LLMs such as GPT-4, along with extensive prompt engineering techniques. This reliance stems from the scarcity of high-quality training datasets for optimization modeling, resulting in elevated costs, prolonged processing times, and privacy concerns. To address these challenges, our work is the first to propose a viable path for training open-source LLMs that are capable of optimization modeling and developing solver codes, eventually leading to a superior ability for automating optimization modeling and solving. Particularly, we introduce OR-Instruct, a semi-automated data synthesis framework for optimization modeling that enables customizable enhancements for specific scenarios or model types. We also introduce IndustryOR, the first industrial benchmark for evaluating LLMs in solving practical OR problems. We train several 7B-scale open-source LLMs using synthesized data (dubbed ORLMs{https://github.com/Cardinal-Operations/ORLM}), which exhibit significantly enhanced optimization modeling capabilities, achieving state-of-the-art performance across the NL4OPT, MAMO, and IndustryOR benchmarks. Additionally, our experiments highlight the potential of scaling law and reinforcement learning to further enhance the performance of ORLMs. The workflows and human-machine interaction paradigms of ORLMs in practical industrial applications are also discussed in the paper.
Authors: Ryumei Nakada, Yichen Xu, Lexin Li, Linjun Zhang
Abstract: Imbalanced classification and spurious correlation are common challenges in data science and machine learning. Both issues are linked to data imbalance, with certain groups of data samples significantly underrepresented, which in turn would compromise the accuracy, robustness and generalizability of the learned models. Recent advances have proposed leveraging the flexibility and generative capabilities of large language models (LLMs), typically built on transformer architectures, to generate synthetic samples and to augment the observed data. In the context of imbalanced data, LLMs are used to oversample underrepresented groups and have shown promising improvements. However, there is a clear lack of theoretical understanding of such synthetic data approaches. In this article, we develop novel theoretical foundations to systematically study the roles of synthetic samples in addressing imbalanced classification and spurious correlation. Specifically, we first explicitly quantify the benefits of synthetic oversampling. Next, we analyze the scaling dynamics in synthetic data augmentation, and derive the corresponding scaling law. Finally, we demonstrate the capacity of transformer models to generate high-quality synthetic samples. We further conduct extensive numerical experiments to validate the efficacy of the LLM-based synthetic oversampling and augmentation.
Authors: Sebastian Neumayer, Fabian Altekr\"uger
Abstract: Theoretical guarantees for the robust solution of inverse problems have important implications for applications. To achieve both guarantees and high reconstruction quality, we propose learning a pixel-based ridge regularizer with a data-dependent and spatially varying regularization strength. For this architecture, we establish the existence of solutions to the associated variational problem and the stability of its solution operator. Further, we prove that the reconstruction forms a maximum-a-posteriori approach. Simulations for biomedical imaging and material sciences demonstrate that the approach yields high-quality reconstructions even if only a small instance-specific training set is available.
Authors: Andrea Carriero, Davide Pettenuzzo, Shubhranshu Shekhar
Abstract: This paper presents a comparative analysis evaluating the accuracy of Large Language Models (LLMs) against traditional macro time series forecasting approaches. In recent times, LLMs have surged in popularity for forecasting due to their ability to capture intricate patterns in data and quickly adapt across very different domains. However, their effectiveness in forecasting macroeconomic time series data compared to conventional methods remains an area of interest. To address this, we conduct a rigorous evaluation of LLMs against traditional macro forecasting methods, using as common ground the FRED-MD database. Our findings provide valuable insights into the strengths and limitations of LLMs in forecasting macroeconomic time series, shedding light on their applicability in real-world scenarios
Authors: Alan J. X. Guo, Mengyi Wei, Yufan Dai, Yali Wei, Pengchen Zhang
Abstract: Insertion, deletion, and substitution (IDS) error-correcting codes have garnered increased attention with recent advancements in DNA storage technology. However, a universal method for designing IDS-correcting codes across varying channel settings remains underexplored. We present an autoencoder-based method, THEA-code, aimed at efficiently generating IDS-correcting codes for complex IDS channels. In the work, a Gumbel-Softmax discretization constraint is proposed to discretize the features of the autoencoder, and a simulated differentiable IDS channel is developed as a differentiable alternative for IDS operations. These innovations facilitate the successful convergence of the autoencoder, resulting in channel-customized IDS-correcting codes with commendable performance across complex IDS channels.
Authors: Justin Bunker, Mark Girolami, Hefin Lambley, Andrew M. Stuart, T. J. Sullivan
Abstract: Autoencoders have found widespread application in both their original deterministic form and in their variational formulation (VAEs). In scientific applications and in image processing it is often of interest to consider data that are viewed as functions; while discretisation (of differential equations arising in the sciences) or pixellation (of images) renders problems finite dimensional in practice, conceiving first of algorithms that operate on functions, and only then discretising or pixellating, leads to better algorithms that smoothly operate between resolutions. In this paper function-space versions of the autoencoder (FAE) and variational autoencoder (FVAE) are introduced, analysed, and deployed. Well-definedness of the objective governing VAEs is a subtle issue, particularly in function space, limiting applicability. For the FVAE objective to be well defined requires compatibility of the data distribution with the chosen generative model; this can be achieved, for example, when the data arise from a stochastic differential equation, but is generally restrictive. The FAE objective, on the other hand, is well defined in many situations where FVAE fails to be. Pairing the FVAE and FAE objectives with neural operator architectures that can be evaluated on any mesh enables new applications of autoencoders to inpainting, superresolution, and generative modelling of scientific data.
Authors: Philip Wiese, Gamze \.Islamo\u{g}lu, Moritz Scherer, Luka Macan, Victor J. B. Jung, Alessio Burrello, Francesco Conti, Luca Benini
Abstract: One of the challenges for Tiny Machine Learning (tinyML) is keeping up with the evolution of Machine Learning models from Convolutional Neural Networks to Transformers. We address this by leveraging a heterogeneous architectural template coupling RISC-V processors with hardwired accelerators supported by an automated deployment flow. We demonstrate Attention-based models in a tinyML power envelope with an octa-core cluster coupled with an accelerator for quantized Attention. Our deployment flow enables end-to-end 8-bit Transformer inference, achieving leading-edge energy efficiency and throughput of 2960 GOp/J and 154 GOp/s (0.65 V, 22 nm FD-SOI technology).
Authors: Zhanhao Zhao, Shaofeng Cai, Haotian Gao, Hexiang Pan, Siqi Xiang, Naili Xing, Gang Chen, Beng Chin Ooi, Yanyan Shen, Yuncheng Wu, Meihui Zhang
Abstract: Databases are increasingly embracing AI to provide autonomous system optimization and intelligent in-database analytics, aiming to relieve end-user burdens across various industry sectors. Nonetheless, most existing approaches fail to account for the dynamic nature of databases, which renders them ineffective for real-world applications characterized by evolving data and workloads. This paper introduces NeurDB, an AI-powered autonomous database that deepens the fusion of AI and databases with adaptability to data and workload drift. NeurDB establishes a new in-database AI ecosystem that seamlessly integrates AI workflows within the database. This integration enables efficient and effective in-database AI analytics and fast-adaptive learned system components. Empirical evaluations demonstrate that NeurDB substantially outperforms existing solutions in managing AI analytics tasks, with the proposed learned components more effectively handling environmental dynamism than state-of-the-art approaches.
Authors: Chao Shen, Wenkang Zhan, Kaiyao Xin, Shujie Pan, Xiaotian Cheng, Ruixiang Liu, Zhe Feng, Chaoyuan Jin, Hui Cong, Chi Xu, Bo Xu, Tien Khee Ng, Siming Chen, Chunlai Xue, Zhanguo Wang, Chao Zhao
Abstract: The semiconductor industry has prioritized automating repetitive tasks through closed-loop, self-driving experimentation, accelerating the optimization of complex multi-step processes. The emergence of machine learning (ML) has ushered in self-driving processes with minimal human intervention. This work introduces SemiEpi, a self-driving platform designed to execute molecular beam epitaxy (MBE) growth of semiconductor heterostructures through multi-step processes, in-situ monitoring, and on-the-fly feedback control. By integrating standard reactor, parameter initialization, and multiple ML models, SemiEpi identifies optimal initial conditions and proposes experiments for multi-step heterostructure growth, eliminating the need for extensive expertise in MBE processes. SemiEpi initializes material growth parameters tailored to specific material characteristics, and fine-tuned control over the growth process is then achieved through ML optimization. We optimize the growth for InAs quantum dots (QDs) heterostructures to showcase the power of SemiEpi, achieving a QD density of 5E10/cm2, 1.6-fold increased photoluminescence (PL) intensity and reduced full width at half maximum (FWHM) of 29.13 meV. This work highlights the potential of closed-loop, ML-guided systems to address challenges in multi-step growth. Our method is critical to achieve repeatable materials growth using commercially scalable tools. Furthermore, our strategy facilitates developing a hardware-independent process and enhancing process repeatability and stability, even without exhaustive knowledge of growth parameters.
Authors: Amirhosein Chahe, Lifeng Zhou
Abstract: This paper introduces a novel method for open-vocabulary 3D scene querying in autonomous driving by combining Language Embedded 3D Gaussians with Large Language Models (LLMs). We propose utilizing LLMs to generate both contextually canonical phrases and helping positive words for enhanced segmentation and scene interpretation. Our method leverages GPT-3.5 Turbo as an expert model to create a high-quality text dataset, which we then use to fine-tune smaller, more efficient LLMs for on-device deployment. Our comprehensive evaluation on the WayveScenes101 dataset demonstrates that LLM-guided segmentation significantly outperforms traditional approaches based on predefined canonical phrases. Notably, our fine-tuned smaller models achieve performance comparable to larger expert models while maintaining faster inference times. Through ablation studies, we discover that the effectiveness of helping positive words correlates with model scale, with larger models better equipped to leverage additional semantic information. This work represents a significant advancement towards more efficient, context-aware autonomous driving systems, effectively bridging 3D scene representation with high-level semantic querying while maintaining practical deployment considerations.
Authors: Martin Bauer, Facundo M\'emoli, Tom Needham, Mao Nishino
Abstract: The Gromov-Wasserstein (GW) distance is a powerful tool for comparing metric measure spaces which has found broad applications in data science and machine learning. Driven by the need to analyze datasets whose objects have increasingly complex structure (such as node and edge-attributed graphs), several variants of GW distance have been introduced in the recent literature. With a view toward establishing a general framework for the theory of GW-like distances, this paper considers a vast generalization of the notion of a metric measure space: for an arbitrary metric space $Z$, we define a $Z$-network to be a measure space endowed with a kernel valued in $Z$. We introduce a method for comparing $Z$-networks by defining a generalization of GW distance, which we refer to as $Z$-Gromov-Wasserstein ($Z$-GW) distance. This construction subsumes many previously known metrics and offers a unified approach to understanding their shared properties. This paper demonstrates that the $Z$-GW distance defines a metric on the space of $Z$-networks which retains desirable properties of $Z$, such as separability, completeness, and geodesicity. Many of these properties were unknown for existing variants of GW distance that fall under our framework. Our focus is on foundational theory, but our results also include computable lower bounds and approximations of the distance which will be useful for practical applications.
Authors: Xinyuan Zheng, Orren Ravid, Robert A. J. Barry, Yoojean Kim, Qian Wang, Young-geun Kim, Xi Zhu, Xiaofu He
Abstract: Autism spectrum disorders (ASDs) are developmental conditions characterized by restricted interests and difficulties in communication. The complexity of ASD has resulted in a deficiency of objective diagnostic biomarkers. Deep learning methods have gained recognition for addressing these challenges in neuroimaging analysis, but finding and interpreting such diagnostic biomarkers are still challenging computationally. Here, we propose a feature reduction pipeline using resting-state fMRI data. We used Craddock atlas and Power atlas to extract functional connectivity data from rs-fMRI, resulting in over 30 thousand features. By using a denoising variational autoencoder, our proposed pipeline further compresses the connectivity features into 5 latent Gaussian distributions, providing is a low-dimensional representation of the data to promote computational efficiency and interpretability. To test the method, we employed the extracted latent representations to classify ASD using traditional classifiers such as SVM on a large multi-site dataset. The 95% confidence interval for the prediction accuracy of SVM is [0.63, 0.76] after site harmonization using the extracted latent distributions. Without using DVAE for dimensionality reduction, the prediction accuracy is 0.70, which falls within the interval. The DVAE successfully encoded the diagnostic information from rs-fMRI data without sacrificing prediction performance. The runtime for training the DVAE and obtaining classification results from its extracted latent features was 7 times shorter compared to training classifiers directly on the raw data. Our findings suggest that the Power atlas provides more effective brain connectivity insights for diagnosing ASD than Craddock atlas. Additionally, we visualized the latent representations to gain insights into the brain networks contributing to the differences between ASD and neurotypical brains.
Authors: Qianyi Deng, Oishi Deb, Amir Patel, Christian Rupprecht, Philip Torr, Niki Trigoni, Andrew Markham
Abstract: Animal pose estimation (APE) aims to locate the animal body parts using a diverse array of sensor and modality inputs (e.g. RGB cameras, LiDAR, infrared, IMU, acoustic and language cues), which is crucial for research across neuroscience, biomechanics, and veterinary medicine. By evaluating 176 papers since 2011, APE methods are categorised by their input sensor and modality types, output forms, learning paradigms, experimental setup, and application domains, presenting detailed analyses of current trends, challenges, and future directions in single- and multi-modality APE systems. The analysis also highlights the transition between human and animal pose estimation, and how innovations in APE can reciprocally enrich human pose estimation and the broader machine learning paradigm. Additionally, 2D and 3D APE datasets and evaluation metrics based on different sensors and modalities are provided. A regularly updated project page is provided here: https://github.com/ChennyDeng/MM-APE.
Authors: Yang Zhang, Fa Wang, Xin Huang, Xintao Li, Sibei Liu, Hansong Zhang
Abstract: This study develops a cloud-based deep learning system for early prediction of diabetes, leveraging the distributed computing capabilities of the AWS cloud platform and deep learning technologies to achieve efficient and accurate risk assessment. The system utilizes EC2 p3.8xlarge GPU instances to accelerate model training, reducing training time by 93.2% while maintaining a prediction accuracy of 94.2%. With an automated data processing and model training pipeline built using Apache Airflow, the system can complete end-to-end updates within 18.7 hours. In clinical applications, the system demonstrates a prediction accuracy of 89.8%, sensitivity of 92.3%, and specificity of 95.1%. Early interventions based on predictions lead to a 37.5% reduction in diabetes incidence among the target population. The system's high performance and scalability provide strong support for large-scale diabetes prevention and management, showcasing significant public health value.
Authors: Shrey Vishen, Jatin Sarabu, Saurav Kumar, Chinmay Bharathulwar, Rithwick Lakshmanan, Vishnu Srinivas
Abstract: We present a novel method for diffusion-guided frameworks for view-consistent super-resolution (SR) in neural rendering. Our approach leverages existing 2D SR models in conjunction with advanced techniques such as Variational Score Distilling (VSD) and a LoRA fine-tuning helper, with spatial training to significantly boost the quality and consistency of upscaled 2D images compared to the previous methods in the literature, such as Renoised Score Distillation (RSD) proposed in DiSR-NeRF (1), or SDS proposed in DreamFusion. The VSD score facilitates precise fine-tuning of SR models, resulting in high-quality, view-consistent images. To address the common challenge of inconsistencies among independent SR 2D images, we integrate Iterative 3D Synchronization (I3DS) from the DiSR-NeRF framework. Our quantitative benchmarks and qualitative results on the LLFF dataset demonstrate the superior performance of our system compared to existing methods such as DiSR-NeRF.
Authors: Naren Sengodan
Abstract: Breast cancer histopathology image classification is critical for early detection and improved patient outcomes. 1 This study introduces a novel approach leveraging EfficientNetV2 models, to improve feature extraction and focus on relevant tissue regions. The proposed models were evaluated on the BreakHis dataset across multiple magnification scales (40X, 100X, 200X, and 400X). 2 Among them, the EfficientNetV2-XL with CBAM achieved outstanding performance, reaching a peak accuracy of 98.96 percent and an F1-score of 98.31 percent at 400X magnification, outperforming state-of-the-art methods. 3 By integrating Contrast Limited Adaptive Histogram Equalization (CLAHE) for preprocessing and optimizing computational efficiency, this method demonstrates its suitability for real-time clinical deployment. 3 The results underscore the potential of attention-enhanced scalable architectures in advancing diagnostic precision for breast cancer detection.
Authors: Longbiao Cheng, Ashutosh Pandey, Buye Xu, Tobi Delbruck, Vamsi Krishna Ithapu, Shih-Chii Liu
Abstract: Deep learning-based speech enhancement (SE) methods often face significant computational challenges when needing to meet low-latency requirements because of the increased number of frames to be processed. This paper introduces the SlowFast framework which aims to reduce computation costs specifically when low-latency enhancement is needed. The framework consists of a slow branch that analyzes the acoustic environment at a low frame rate, and a fast branch that performs SE in the time domain at the needed higher frame rate to match the required latency. Specifically, the fast branch employs a state space model where its state transition process is dynamically modulated by the slow branch. Experiments on a SE task with a 2 ms algorithmic latency requirement using the Voice Bank + Demand dataset show that our approach reduces computation cost by 70% compared to a baseline single-branch network with equivalent parameters, without compromising enhancement performance. Furthermore, by leveraging the SlowFast framework, we implemented a network that achieves an algorithmic latency of just 62.5 {\mu}s (one sample point at 16 kHz sample rate) with a computation cost of 100 M MACs/s, while scoring a PESQ-NB of 3.12 and SISNR of 16.62.
Authors: Shrihan Agarwal, Aleksandra \'Ciprijanovi\'c, Brian D. Nord
Abstract: Modeling strong gravitational lenses is computationally expensive for the complex data from modern and next-generation cosmic surveys. Deep learning has emerged as a promising approach for finding lenses and predicting lensing parameters, such as the Einstein radius. Mean-variance Estimators (MVEs) are a common approach for obtaining aleatoric (data) uncertainties from a neural network prediction. However, neural networks have not been demonstrated to perform well on out-of-domain target data successfully - e.g., when trained on simulated data and applied to real, observational data. In this work, we perform the first study of the efficacy of MVEs in combination with unsupervised domain adaptation (UDA) on strong lensing data. The source domain data is noiseless, and the target domain data has noise mimicking modern cosmology surveys. We find that adding UDA to MVE increases the accuracy on the target data by a factor of about two over an MVE model without UDA. Including UDA also permits much more well-calibrated aleatoric uncertainty predictions. Advancements in this approach may enable future applications of MVE models to real observational data.
Authors: Murad Mehrab Abrar, Souryadeep Mondal, Michelle Hickner
Abstract: This paper presents a Sim2Real (Simulation to Reality) approach to bridge the gap between a trained agent in a simulated environment and its real-world implementation in navigating a robot in a similar setting. Specifically, we focus on navigating a quadruped robot in a real-world grid-like environment inspired by the Gymnasium Frozen Lake -- a highly user-friendly and free Application Programming Interface (API) to develop and test Reinforcement Learning (RL) algorithms. We detail the development of a pipeline to transfer motion policies learned in the Frozen Lake simulation to a physical quadruped robot, thus enabling autonomous navigation and obstacle avoidance in a grid without relying on expensive localization and mapping sensors. The work involves training an RL agent in the Frozen Lake environment and utilizing the resulting Q-table to control a 12 Degrees-of-Freedom (DOF) quadruped robot. In addition to detailing the RL implementation, inverse kinematics-based quadruped gaits, and the transfer policy pipeline, we open-source the project on GitHub and include a demonstration video of our Sim2Real transfer approach. This work provides an accessible, straightforward, and low-cost framework for researchers, students, and hobbyists to explore and implement RL-based robot navigation in real-world grid environments.
Authors: Shengqi Chen, Zilin Wang, Jianrong Dai, Shirui Qin, Ying Cao, Ruiao Zhao, Jiayun Chen, Guohua Wu, Yuan Tang
Abstract: Background and Purpose: Accurate motion tracking in MRI-guided Radiotherapy (MRIgRT) is essential for effective treatment delivery. This study aimed to enhance motion tracking precision in MRIgRT through an automatic real-time markerless tracking method using an enhanced Tracking-Learning-Detection (ETLD) framework with automatic segmentation. Materials and Methods: We developed a novel MRIgRT motion tracking and segmentation method by integrating the ETLD framework with an improved Chan-Vese model (ICV), named ETLD+ICV. The ETLD framework was upgraded for real-time cine MRI, including advanced image preprocessing, no-reference image quality assessment, an enhanced median-flow tracker, and a refined detector with dynamic search region adjustments. ICV was used for precise target volume coverage, refining the segmented region frame by frame using tracking results, with key parameters optimized. The method was tested on 3.5D MRI scans from 10 patients with liver metastases. Results: Evaluation of 106,000 frames across 77 treatment fractions showed sub-millimeter tracking errors of less than 0.8mm, with over 99% precision and 98% recall for all subjects in the Beam Eye View(BEV)/Beam Path View(BPV) orientation. The ETLD+ICV method achieved a dice global score of more than 82% for all subjects, demonstrating the method's extensibility and precise target volume coverage. Conclusion: This study successfully developed an automatic real-time markerless motion tracking method for MRIgRT that significantly outperforms current methods. The novel method not only delivers exceptional precision in tracking and segmentation but also shows enhanced adaptability to clinical demands, making it an indispensable asset in improving the efficacy of radiotherapy treatments.
Authors: Ivory Yang, Weicheng Ma, Soroush Vosoughi
Abstract: The preservation and revitalization of endangered and extinct languages is a meaningful endeavor, conserving cultural heritage while enriching fields like linguistics and anthropology. However, these languages are typically low-resource, making their reconstruction labor-intensive and costly. This challenge is exemplified by Nushu, a rare script historically used by Yao women in China for self-expression within a patriarchal society. To address this challenge, we introduce NushuRescue, an AI-driven framework designed to train large language models (LLMs) on endangered languages with minimal data. NushuRescue automates evaluation and expands target corpora to accelerate linguistic revitalization. As a foundational component, we developed NCGold, a 500-sentence Nushu-Chinese parallel corpus, the first publicly available dataset of its kind. Leveraging GPT-4-Turbo, with no prior exposure to Nushu and only 35 short examples from NCGold, NushuRescue achieved 48.69% translation accuracy on 50 withheld sentences and generated NCSilver, a set of 98 newly translated modern Chinese sentences of varying lengths. A sample of both NCGold and NCSilver is included in the Supplementary Materials. Additionally, we developed FastText-based and Seq2Seq models to further support research on Nushu. NushuRescue provides a versatile and scalable tool for the revitalization of endangered languages, minimizing the need for extensive human input.
Authors: Jiahao Cui, Hui Li, Yun Zhan, Hanlin Shang, Kaihui Cheng, Yuqi Ma, Shan Mu, Hang Zhou, Jingdong Wang, Siyu Zhu
Abstract: Existing methodologies for animating portrait images face significant challenges, particularly in handling non-frontal perspectives, rendering dynamic objects around the portrait, and generating immersive, realistic backgrounds. In this paper, we introduce the first application of a pretrained transformer-based video generative model that demonstrates strong generalization capabilities and generates highly dynamic, realistic videos for portrait animation, effectively addressing these challenges. The adoption of a new video backbone model makes previous U-Net-based methods for identity maintenance, audio conditioning, and video extrapolation inapplicable. To address this limitation, we design an identity reference network consisting of a causal 3D VAE combined with a stacked series of transformer layers, ensuring consistent facial identity across video sequences. Additionally, we investigate various speech audio conditioning and motion frame mechanisms to enable the generation of continuous video driven by speech audio. Our method is validated through experiments on benchmark and newly proposed wild datasets, demonstrating substantial improvements over prior methods in generating realistic portraits characterized by diverse orientations within dynamic and immersive scenes. Further visualizations and the source code are available at: https://fudan-generative-vision.github.io/hallo3/.
Authors: Kyriakos Flouris, Anna Volokitin, Gustav Bredell, Ender Konukoglu
Abstract: The autoencoder model typically uses an encoder to map data to a lower dimensional latent space and a decoder to reconstruct it. However, relying on an encoder for inversion can lead to suboptimal representations, particularly limiting in physical sciences where precision is key. We introduce a decoder-only method using gradient flow to directly encode data into the latent space, defined by ordinary differential equations (ODEs). This approach eliminates the need for approximate encoder inversion. We train the decoder via the adjoint method and show that costly integrals can be avoided with minimal accuracy loss. Additionally, we propose a $2^{nd}$ order ODE variant, approximating Nesterov's accelerated gradient descent for faster convergence. To handle stiff ODEs, we use an adaptive solver that prioritizes loss minimization, improving robustness. Compared to traditional autoencoders, our method demonstrates explicit encoding and superior data efficiency, which is crucial for data-scarce scenarios in the physical sciences. Furthermore, this work paves the way for integrating machine learning into scientific workflows, where precise and efficient encoding is critical. \footnote{The code for this work is available at \url{https://github.com/k-flouris/gfe}.}
Authors: Andrew Wagenmaker, Lu Mi, Marton Rozsa, Matthew S. Bull, Karel Svoboda, Kayvon Daie, Matthew D. Golub, Kevin Jamieson
Abstract: Recent advances in techniques for monitoring and perturbing neural populations have greatly enhanced our ability to study circuits in the brain. In particular, two-photon holographic optogenetics now enables precise photostimulation of experimenter-specified groups of individual neurons, while simultaneous two-photon calcium imaging enables the measurement of ongoing and induced activity across the neural population. Despite the enormous space of potential photostimulation patterns and the time-consuming nature of photostimulation experiments, very little algorithmic work has been done to determine the most effective photostimulation patterns for identifying the neural population dynamics. Here, we develop methods to efficiently select which neurons to stimulate such that the resulting neural responses will best inform a dynamical model of the neural population activity. Using neural population responses to photostimulation in mouse motor cortex, we demonstrate the efficacy of a low-rank linear dynamical systems model, and develop an active learning procedure which takes advantage of low-rank structure to determine informative photostimulation patterns. We demonstrate our approach on both real and synthetic data, obtaining in some cases as much as a two-fold reduction in the amount of data required to reach a given predictive power. Our active stimulation design method is based on a novel active learning procedure for low-rank regression, which may be of independent interest.
Authors: Zheng Cao, Helyette Geman
Abstract: This manuscript introduces the Hype-Adjusted Probability Measure developed in the context of a new Natural Language Processing (NLP) approach for stock return and volatility forecasting. A novel sentiment score equation is presented to capture component and memory effects and assign dynamic parameters, enhancing the impact of intraday news data on forecasting next-period volatility for selected U.S. semiconductor tickers. This approach integrates machine learning techniques to analyze and improve the predictive value of news. Building on the research of Geman et al [6], this work improves forecast accuracy by addressing news bias, memory, and weight, and incorporating shifts in senti-ment direction. Finally, we propose the Hype-Adjusted Probability Measure, proving its existence and uniqueness, and discuss its theoretical applications in finance for NLP-based stock return forecasting, outlining future research pathways inspired by its concepts.
Authors: Yu Chen, Shuai Zheng, Nianyi Wang, Menglong Jin, Yan Chang
Abstract: Fluid simulation is an important research topic in computer graphics (CG) and animation in video games. Traditional methods based on Navier-Stokes equations are computationally expensive. In this paper, we treat fluid motion as point cloud transformation and propose the first neural network method specifically designed for efficient and robust fluid simulation in complex environments. This model is also the deep learning model that is the first to be capable of stably modeling fluid particle dynamics in such complex scenarios. Our triangle feature fusion design achieves an optimal balance among fluid dynamics modeling, momentum conservation constraints, and global stability control. We conducted comprehensive experiments on datasets. Compared to existing neural network-based fluid simulation algorithms, we significantly enhanced accuracy while maintaining high computational speed. Compared to traditional SPH methods, our speed improved approximately 10 times. Furthermore, compared to traditional fluid simulation software such as Flow3D, our computation speed increased by more than 300 times.
Authors: Luqi Wang, Wenbao Jiang
Abstract: As a cornerstone of blockchain technology in the 3.0 era, smart contracts play a pivotal role in the evolution of blockchain systems. In order to address the limitations of existing smart contract vulnerability detection models with regard to their generalisation capability, an AF-STip smart contract vulnerability detection framework incorporating efficient knowledge migration is proposed. AF-STip employs the teacher network as the main model and migrates the knowledge processed by the smart contract to the student model using a data-free knowledge distillation method. The student model utilises this knowledge to enhance its vulnerability detection capabilities. The approach markedly enhances the model's capacity for feature extraction and cross-class adaptation, while concurrently reducing computational overhead.In order to further enhance the extraction of vulnerability features, an adaptive fusion module is proposed in this paper, which aims to strengthen the interaction and fusion of feature information.The experimental results demonstrate that the STip model attains an average F1 value detection score of 91.16% for the four vulnerabilities without disclosing the original smart contract data. To validate the viability of the proposed lightweight migration approach, the student model is deployed in a migration learning task targeting a novel vulnerability type, resulting in an accuracy of 91.02% and an F1 score of 90.46%. To the best of our knowledge, AF-STip is the inaugural model to apply data-free knowledge migration to smart contract vulnerability detection. While markedly reducing the computational overhead, the method still demonstrates exceptional performance in detecting novel vulnerabilities.
Authors: Benjamin Steenhoek, Michele Tufano, Neel Sundaresan, Alexey Svyatkovskiy
Abstract: Software testing is a crucial but time-consuming aspect of software development, and recently, Large Language Models (LLMs) have gained popularity for automated test case generation. However, because LLMs are trained on vast amounts of open-source code, they often generate test cases that do not adhere to best practices and may even contain test smells (anti-patterns). To address this issue, we propose Reinforcement Learning from Static Quality Metrics (RLSQM), wherein we utilize Reinforcement Learning to generate high-quality unit tests based on static analysis-based quality metrics. First, we analyzed LLM-generated tests and show that LLMs frequently do generate undesirable test smells -- up to 37% of the time. Then, we implemented lightweight static analysis-based reward model and trained LLMs using this reward model to optimize for five code quality metrics. Our experimental results demonstrate that the RL-optimized Codex model consistently generated higher-quality test cases than the base LLM, improving quality metrics by up to 23%, and generated nearly 100% syntactically-correct code. RLSQM also outperformed GPT-4 on all code quality metrics, in spite of training a substantially cheaper Codex model. We provide insights into how reliably utilize RL to improve test generation quality and show that RLSQM is a significant step towards enhancing the overall efficiency and reliability of automated software testing. Our data are available at https://doi.org/10.6084/m9.figshare.25983166.
Authors: Jinheon Baek, Sun Jae Lee, Prakhar Gupta, Geunseob Oh, Siddharth Dalmia, Prateek Kolhar
Abstract: In-Context Learning (ICL) is a technique by which language models make predictions based on examples provided in their input context. Previously, their context window size imposed a limit on the number of examples that can be shown, making example selection techniques crucial for identifying the maximally effective set of examples. However, the recent advent of Long Context Language Models (LCLMs) has significantly increased the number of examples that can be included in context, raising an important question of whether ICL performance in a many-shot regime is still sensitive to the method of sample selection. To answer this, we revisit these approaches in the context of LCLMs through extensive experiments on 18 datasets spanning 4 tasks. Surprisingly, we observe that sophisticated example selection techniques do not yield significant improvements over a simple random sample selection method. Instead, we find that the advent of LCLMs has fundamentally shifted the challenge of ICL from that of selecting the most effective examples to that of collecting sufficient examples to fill the context window. Specifically, in certain datasets, including all available examples does not fully utilize the context window; however, by augmenting the examples in context with a simple data augmentation approach, we substantially improve ICL performance by 5%.
Authors: Mengxin Wang (Naveen Jindal School of Management, The University of Texas at Dallas), Dennis J. Zhang (Olin School of Business, Washington University in St. Louis), Heng Zhang (W. P. Carey School of Business, Arizona State University)
Abstract: Large Language Models (LLMs) have transformed artificial intelligence by excelling in complex natural language processing tasks. Their ability to generate human-like text has opened new possibilities for market research, particularly in conjoint analysis, where understanding consumer preferences is essential but often resource-intensive. Traditional survey-based methods face limitations in scalability and cost, making LLM-generated data a promising alternative. However, while LLMs have the potential to simulate real consumer behavior, recent studies highlight a significant gap between LLM-generated and human data, with biases introduced when substituting between the two. In this paper, we address this gap by proposing a novel statistical data augmentation approach that efficiently integrates LLM-generated data with real data in conjoint analysis. Our method leverages transfer learning principles to debias the LLM-generated data using a small amount of human data. This results in statistically robust estimators with consistent and asymptotically normal properties, in contrast to naive approaches that simply substitute human data with LLM-generated data, which can exacerbate bias. We validate our framework through an empirical study on COVID-19 vaccine preferences, demonstrating its superior ability to reduce estimation error and save data and costs by 24.9% to 79.8%. In contrast, naive approaches fail to save data due to the inherent biases in LLM-generated data compared to human data. Another empirical study on sports car choices validates the robustness of our results. Our findings suggest that while LLM-generated data is not a direct substitute for human responses, it can serve as a valuable complement when used within a robust statistical framework.
Authors: Huigang Chen, Edwin Ng, Slawek Smyl, Gavin Steininger
Abstract: This paper introduces a recurrent neural network approach for predicting user lifetime value in Software as a Service (SaaS) applications. The approach accounts for three connected time dimensions. These dimensions are the user cohort (the date the user joined), user age-in-system (the time since the user joined the service) and the calendar date the user is an age-in-system (i.e., contemporaneous information).The recurrent neural networks use a multi-cell architecture, where each cell resembles a long short-term memory neural network. The approach is applied to predicting both acquisition (new users) and rolling (existing user) lifetime values for a variety of time horizons. It is found to significantly improve median absolute percent error versus light gradient boost models and Buy Until You Die models.
Authors: Zibin Pan, Shuwen Zhang, Yuesheng Zheng, Chi Li, Yuheng Cheng, Junhua Zhao
Abstract: Machine unlearning in the domain of large language models (LLMs) has attracted great attention recently, which aims to effectively eliminate undesirable behaviors from LLMs without full retraining from scratch. In this paper, we explore the Gradient Ascent (GA) approach in LLM unlearning, which is a proactive way to decrease the prediction probability of the model on the target data in order to remove their influence. We analyze two challenges that render the process impractical: gradient explosion and catastrophic forgetting. To address these issues, we propose Multi-Objective Large Language Model Unlearning (MOLLM) algorithm. We first formulate LLM unlearning as a multi-objective optimization problem, in which the cross-entropy loss is modified to the unlearning version to overcome the gradient explosion issue. A common descent update direction is then calculated, which enables the model to forget the target data while preserving the utility of the LLM. Our empirical results verify that MoLLM outperforms the SOTA GA-based LLM unlearning methods in terms of unlearning effect and model utility preservation. The source code is available at https://github.com/zibinpan/MOLLM.
Authors: Grace Jiang, Jiangchao Qiu, Sai Ravela
Abstract: Identifying tropical cyclones that generate destructive storm tides for risk assessment, such as from large downscaled storm catalogs for climate studies, is often intractable because it entails many expensive Monte Carlo hydrodynamic simulations. Here, we show that surrogate models are promising from accuracy, recall, and precision perspectives, and they "generalize" to novel climate scenarios. We then present an informative online learning approach to rapidly search for extreme storm tide-producing cyclones using only a few hydrodynamic simulations. Starting from a minimal subset of TCs with detailed storm tide hydrodynamic simulations, a surrogate model selects informative data to retrain online and iteratively improves its predictions of damaging TCs. Results on an extensive catalog of downscaled TCs indicate 100% precision in retrieving rare destructive storms using less than 20% of the simulations as training. The informative sampling approach is efficient, scalable to large storm catalogs, and generalizable to climate scenarios.
Authors: Suttisak Wizadwongsa, Jinfan Zhou, Edward Li, Jeong Joon Park
Abstract: Recent AI-based 3D content creation has largely evolved along two paths: feed-forward image-to-3D reconstruction approaches and 3D generative models trained with 2D or 3D supervision. In this work, we show that existing feed-forward reconstruction methods can serve as effective latent encoders for training 3D generative models, thereby bridging these two paradigms. By reusing powerful pre-trained reconstruction models, we avoid computationally expensive encoder network training and obtain rich 3D latent features for generative modeling for free. However, the latent spaces of reconstruction models are not well-suited for generative modeling due to their unstructured nature. To enable flow-based model training on these latent features, we develop post-processing pipelines, including protocols to standardize the features and spatial weighting to concentrate on important regions. We further incorporate a 2D image space perceptual rendering loss to handle the high-dimensional latent spaces. Finally, we propose a multi-stream transformer-based rectified flow architecture to achieve linear scaling and high-quality text-conditioned 3D generation. Our framework leverages the advancements of feed-forward reconstruction models to enhance the scalability of 3D generative modeling, achieving both high computational efficiency and state-of-the-art performance in text-to-3D generation.
Authors: Xindi Wu, Mengzhou Xia, Rulin Shao, Zhiwei Deng, Pang Wei Koh, Olga Russakovsky
Abstract: Visual Instruction Tuning typically requires a large amount of vision-language training data. This data often containing redundant information that increases computational costs without proportional performance gains. In this work, we introduce ICONS, a gradient-driven Influence CONsensus approach for vision-language data Selection that selects a compact training dataset for efficient multi-task training. The key element of our approach is cross-task influence consensus, which uses majority voting across task-specific influence matrices to identify samples that are consistently valuable across multiple tasks, allowing us to effectively prioritize data that optimizes for overall performance. Experiments show that models trained on our selected data (20% of LLaVA-665K) achieve 98.6% of the relative performance obtained using the full dataset. Additionally, we release this subset, LLaVA-ICONS-133K, a compact yet highly informative subset of LLaVA-665K visual instruction tuning data, preserving high impact training data for efficient vision-language model development.
Authors: Phuc Nguyen, Miao Li, Alexandra Morgan, Rima Arnaout, Ramy Arnaout
Abstract: Generative models hold great potential, but only if one can trust the evaluation of the data they generate. We show that many commonly used quality scores for comparing two-dimensional distributions of synthetic vs. ground-truth data give better results than they should, a phenomenon we call the "grade inflation problem." We show that the correlation score, Jaccard score, earth-mover's score, and Kullback-Leibler (relative-entropy) score all suffer grade inflation. We propose that any score that values all datapoints equally, as these do, will also exhibit grade inflation; we refer to such scores as "equipoint" scores. We introduce the concept of "equidensity" scores, and present the Eden score, to our knowledge the first example of such a score. We found that Eden avoids grade inflation and agrees better with human perception of goodness-of-fit than the equipoint scores above. We propose that any reasonable equidensity score will avoid grade inflation. We identify a connection between equidensity scores and R\'enyi entropy of negative order. We conclude that equidensity scores are likely to outperform equipoint scores for generative models, and for comparing low-dimensional distributions more generally.
Authors: Yihan He, Yuan Cao, Hong-Yu Chen, Dennis Wu, Jianqing Fan, Han Liu
Abstract: Transformers demonstrate significant advantages as the building block of modern LLMs. In this work, we study the capacities of Transformers in performing unsupervised learning. We show that multi-layered Transformers, given a sufficiently large set of pre-training instances, are able to learn the algorithms themselves and perform statistical estimation tasks given new instances. This learning paradigm is distinct from the in-context learning setup and is similar to the learning procedure of human brains where skills are learned through past experience. Theoretically, we prove that pre-trained Transformers can learn the spectral methods and use the classification of bi-class Gaussian mixture model as an example. Our proof is constructive using algorithmic design techniques. Our results are built upon the similarities of multi-layered Transformer architecture with the iterative recovery algorithms used in practice. Empirically, we verify the strong capacity of the multi-layered (pre-trained) Transformer on unsupervised learning through the lens of both the PCA and the Clustering tasks performed on the synthetic and real-world datasets.
Authors: Jingfeng Yao, Xinggang Wang
Abstract: Latent diffusion models with Transformer architectures excel at generating high-fidelity images. However, recent studies reveal an optimization dilemma in this two-stage design: while increasing the per-token feature dimension in visual tokenizers improves reconstruction quality, it requires substantially larger diffusion models and more training iterations to achieve comparable generation performance. Consequently, existing systems often settle for sub-optimal solutions, either producing visual artifacts due to information loss within tokenizers or failing to converge fully due to expensive computation costs. We argue that this dilemma stems from the inherent difficulty in learning unconstrained high-dimensional latent spaces. To address this, we propose aligning the latent space with pre-trained vision foundation models when training the visual tokenizers. Our proposed VA-VAE (Vision foundation model Aligned Variational AutoEncoder) significantly expands the reconstruction-generation frontier of latent diffusion models, enabling faster convergence of Diffusion Transformers (DiT) in high-dimensional latent spaces. To exploit the full potential of VA-VAE, we build an enhanced DiT baseline with improved training strategies and architecture designs, termed LightningDiT. The integrated system achieves state-of-the-art (SOTA) performance on ImageNet 256x256 generation with an FID score of 1.35 while demonstrating remarkable training efficiency by reaching an FID score of 2.11 in just 64 epochs--representing an over 21 times convergence speedup compared to the original DiT. Models and codes are available at: https://github.com/hustvl/LightningDiT.
Authors: Shvetank Prakash, Andrew Cheng, Jason Yik, Arya Tschand, Radhika Ghosal, Ikechukwu Uchendu, Jessica Quaye, Jeffrey Ma, Shreyas Grampurohit, Sofia Giannuzzi, Arnav Balyan, Fin Amin, Aadya Pipersenia, Yash Choudhary, Ankita Nayak, Amir Yazdanbakhsh, Vijay Janapa Reddi
Abstract: We introduce QuArch, a dataset of 1500 human-validated question-answer pairs designed to evaluate and enhance language models' understanding of computer architecture. The dataset covers areas including processor design, memory systems, and performance optimization. Our analysis highlights a significant performance gap: the best closed-source model achieves 84% accuracy, while the top small open-source model reaches 72%. We observe notable struggles in memory systems, interconnection networks, and benchmarking. Fine-tuning with QuArch improves small model accuracy by up to 8%, establishing a foundation for advancing AI-driven computer architecture research. The dataset and leaderboard are at https://harvard-edge.github.io/QuArch/.