Authors: Mingzhu Liu, Angela H. Chen, George H. Chen
Abstract: Time series foundation models are pre-trained on large datasets and are able to achieve state-of-the-art performance in diverse tasks. However, to date, there has been limited work demonstrating how well these models perform in medical applications, where labeled data can be scarce. Further, we observe that currently, the majority of time series foundation models either are univariate in nature, or assume channel independence, meaning that they handle multivariate time series but do not model how the different variables relate. In this paper, we propose a prompt-tuning-inspired fine-tuning technique, Generalized Prompt Tuning (Gen-P-Tuning), that enables us to adapt an existing univariate time series foundation model (treated as frozen) to handle multivariate time series prediction. Our approach provides a way to combine information across channels (variables) of multivariate time series. We demonstrate the effectiveness of our fine-tuning approach against various baselines on two MIMIC classification tasks, and on influenza-like illness forecasting.
Authors: Shang Liu, Yu Pan, Guanting Chen, Xiaocheng Li
Abstract: Learning a reward model (RM) from human preferences has been an important component in aligning large language models (LLMs). The canonical setup of learning RMs from pairwise preference data is rooted in the classic Bradley-Terry (BT) model that accepts binary feedback, i.e., the label being either Response 1 is better than Response 2, or the opposite. Such a setup inevitably discards potentially useful samples (such as "tied" between the two responses) and loses more fine-grained information (such as "slightly better"). In this paper, we propose a framework for learning RMs under ordinal feedback which generalizes the case of binary preference feedback to any arbitrary granularity. Specifically, we first identify a marginal unbiasedness condition, which generalizes the assumption of the BT model in the existing binary feedback setting. The condition validates itself via the sociological concept of the wisdom of the crowd. Under the condition, we develop a natural probability model for pairwise preference data under ordinal feedback and analyze its properties. We prove the statistical benefits of ordinal feedback in terms of reducing the Rademacher complexity compared to the case of binary feedback. The proposed learning objective and the theory also extend to hinge loss and direct policy optimization (DPO). In particular, the theoretical analysis may be of independent interest when applying to a seemingly unrelated problem of knowledge distillation to interpret the bias-variance trade-off therein. The framework also sheds light on writing guidance for human annotators. Our numerical experiments validate that fine-grained feedback leads to better reward learning for both in-distribution and out-of-distribution settings. Further experiments show that incorporating a certain proportion of samples with tied preference boosts RM learning.
Authors: Mariette Dupuy, Marie Chavent, Remi Dubois
Abstract: This paper introduces a methodology based on Denoising AutoEncoder (DAE) for missing data imputation. The proposed methodology, called mDAE hereafter, results from a modification of the loss function and a straightforward procedure for choosing the hyper-parameters. An ablation study shows on several UCI Machine Learning Repository datasets, the benefit of using this modified loss function and an overcomplete structure, in terms of Root Mean Squared Error (RMSE) of reconstruction. This numerical study is completed by comparing the mDAE methodology with eight other methods (four standard and four more recent). A criterion called Mean Distance to Best (MDB) is proposed to measure how a method performs globally well on all datasets. This criterion is defined as the mean (over the datasets) of the distances between the RMSE of the considered method and the RMSE of the best method. According to this criterion, the mDAE methodology was consistently ranked among the top methods (along with SoftImput and missForest), while the four more recent methods were systematically ranked last. The Python code of the numerical study will be available on GitHub so that results can be reproduced or generalized with other datasets and methods.
Authors: Poorya Khajouie, Titli Sarkar, Krishna Rauniyar, Li Chen, Wu Xu, Vijay Raghavan
Abstract: Protein structures represent the key to deciphering biological functions. The more detailed form of similarity among these proteins is sometimes overlooked by the conventional structural comparison methods. In contrast, further advanced methods, such as Triangular Spatial Relationship (TSR), have been demonstrated to make finer differentiations. Still, the classical implementation of TSR does not provide for the integration of secondary structure information, which is important for a more detailed understanding of the folding pattern of a protein. To overcome these limitations, we developed the SSE-TSR approach. The proposed method integrates secondary structure elements (SSEs) into TSR-based protein representations. This allows an enriched representation of protein structures by considering 18 different combinations of helix, strand, and coil arrangements. Our results show that using SSEs improves the accuracy and reliability of protein classification to varying degrees. We worked with two large protein datasets of 9.2K and 7.8K samples, respectively. We applied the SSE-TSR approach and used a neural network model for classification. Interestingly, introducing SSEs improved performance statistics for Dataset 1, with accuracy moving from 96.0% to 98.3%. For Dataset 2, where the performance statistics were already good, further small improvements were found with the introduction of SSE, giving an accuracy of 99.5% compared to 99.4%. These results show that SSE integration can dramatically improve TSR key discrimination, with significant benefits in datasets with low initial accuracies and only incremental gains in those with high baseline performance. Thus, SSE-TSR is a powerful bioinformatics tool that improves protein classification and understanding of protein function and interaction.
Authors: Jan Dubi\'nski, Antoni Kowalczuk, Franziska Boenisch, Adam Dziedzic
Abstract: Diffusion Models (DMs) benefit from large and diverse datasets for their training. Since this data is often scraped from the Internet without permission from the data owners, this raises concerns about copyright and intellectual property protections. While (illicit) use of data is easily detected for training samples perfectly re-created by a DM at inference time, it is much harder for data owners to verify if their data was used for training when the outputs from the suspect DM are not close replicas. Conceptually, membership inference attacks (MIAs), which detect if a given data point was used during training, present themselves as a suitable tool to address this challenge. However, we demonstrate that existing MIAs are not strong enough to reliably determine the membership of individual images in large, state-of-the-art DMs. To overcome this limitation, we propose CDI, a framework for data owners to identify whether their dataset was used to train a given DM. CDI relies on dataset inference techniques, i.e., instead of using the membership signal from a single data point, CDI leverages the fact that most data owners, such as providers of stock photography, visual media companies, or even individual artists, own datasets with multiple publicly exposed data points which might all be included in the training of a given DM. By selectively aggregating signals from existing MIAs and using new handcrafted methods to extract features for these datasets, feeding them to a scoring model, and applying rigorous statistical testing, CDI allows data owners with as little as 70 data points to identify with a confidence of more than 99% whether their data was used to train a given DM. Thereby, CDI represents a valuable tool for data owners to claim illegitimate use of their copyrighted data.
Authors: Roberto Dias Algarte
Abstract: This article introduces a novel approach to the mathematical development of Ordinary Least Squares and Neural Network regression models, diverging from traditional methods in current Machine Learning literature. By leveraging Tensor Analysis and fundamental matrix computations, the theoretical foundations of both models are meticulously detailed and extended to their complete algorithmic forms. The study culminates in the presentation of three algorithms, including a streamlined version of the Backpropagation Algorithm for Neural Networks, illustrating the benefits of this new mathematical approach.
Authors: Yucheng Xing, Xin Wang
Abstract: Convolutional Neural Network (CNN) has been applied to more and more scenarios due to its excellent performance in many machine learning tasks, especially with deep and complex structures. However, as the network goes deeper, more parameters need to be stored and optimized. Besides, almost all common CNN models adopt "train-and-use" strategy where the structure is pre-defined and the kernel parameters are fixed after the training with the same structure and set of parameters used for all data without considering the content complexity. In this paper, we propose a new CNN framework, named as $\textit{Puppet-CNN}$, which contains two modules: a $\textit{puppet module}$ and a $\textit{puppeteer module}$. The puppet module is a CNN model used to actually process the input data just like other works, but its depth and kernels are generated by the puppeteer module (realized with Ordinary Differential Equation (ODE)) based on the input complexity each time. By recurrently generating kernel parameters in the puppet module, we can take advantage of the dependence among kernels of different convolutional layers to significantly reduce the size of CNN model by only storing and training the parameters of the much smaller puppeteer ODE module. Through experiments on several datasets, our method has proven to be superior than the traditional CNNs on both performance and efficiency. The model size can be reduced more than 10 times.
Authors: Xuechen Zhang, Xiangyu Chang, Mingchen Li, Amit Roy-Chowdhury, Jiasi Chen, Samet Oymak
Abstract: The attention mechanism within the transformer architecture enables the model to weigh and combine tokens based on their relevance to the query. While self-attention has enjoyed major success, it notably treats all queries $q$ in the same way by applying the mapping $V^\top\text{softmax}(Kq)$, where $V,K$ are the value and key embeddings respectively. In this work, we argue that this uniform treatment hinders the ability to control contextual sparsity and relevance. As a solution, we introduce the $\textit{Selective Self-Attention}$ (SSA) layer that augments the softmax nonlinearity with a principled temperature scaling strategy. By controlling temperature, SSA adapts the contextual sparsity of the attention map to the query embedding and its position in the context window. Through theory and experiments, we demonstrate that this alleviates attention dilution, aids the optimization process, and enhances the model's ability to control softmax spikiness of individual queries. We also incorporate temperature scaling for value embeddings and show that it boosts the model's ability to suppress irrelevant/noisy tokens. Notably, SSA is a lightweight method which introduces less than 0.5% new parameters through a weight-sharing strategy and can be fine-tuned on existing LLMs. Extensive empirical evaluations demonstrate that SSA-equipped models achieve a noticeable and consistent accuracy improvement on language modeling benchmarks.
Authors: Colverd Grace, Schade Laura, Takami Jumpei, Bot Karol, Gallego Joseph
Abstract: Tree species classification plays an important role in nature conservation, forest inventories, forest management, and the protection of endangered species. Over the past four decades, remote sensing technologies have been extensively utilized for tree species classification, with Synthetic Aperture Radar (SAR) emerging as a key technique. In this study, we employed TomoSense, a 3D tomographic dataset, which utilizes a stack of single-look complex (SLC) images, a byproduct of SAR, captured at different incidence angles to generate a three-dimensional representation of the terrain. Our research focuses on evaluating multiple tabular machine-learning models using the height information derived from the tomographic image intensities to classify eight distinct tree species. The SLC data and tomographic imagery were analyzed across different polarimetric configurations and geosplit configurations. We investigated the impact of these variations on classification accuracy, comparing the performance of various tabular machine-learning models and optimizing them using Bayesian optimization. Additionally, we incorporated a proxy for actual tree height using point cloud data from Light Detection and Ranging (LiDAR) to provide height statistics associated with the model's predictions. This comparison offers insights into the reliability of tomographic data in predicting tree species classification based on height.
Authors: Qin Tian, Chen Zhao, Minglai Shao, Wenjun Wang, Yujie Lin, Dong Li
Abstract: Domain generalization on graphs aims to develop models with robust generalization capabilities, ensuring effective performance on the testing set despite disparities between testing and training distributions. However, existing methods often rely on static encoders directly applied to the target domain, constraining its flexible adaptability. In contrast to conventional methodologies, which concentrate on developing specific generalized models, our framework, MLDGG, endeavors to achieve adaptable generalization across diverse domains by integrating cross-multi-domain meta-learning with structure learning and semantic identification. Initially, it introduces a generalized structure learner to mitigate the adverse effects of task-unrelated edges, enhancing the comprehensiveness of representations learned by Graph Neural Networks (GNNs) while capturing shared structural information across domains. Subsequently, a representation learner is designed to disentangle domain-invariant semantic and domain-specific variation information in node embedding by leveraging causal reasoning for semantic identification, further enhancing generalization. In the context of meta-learning, meta-parameters for both learners are optimized to facilitate knowledge transfer and enable effective adaptation to graphs through fine-tuning within the target domains, where target graphs are inaccessible during training. Our empirical results demonstrate that MLDGG surpasses baseline methods, showcasing its effectiveness in three different distribution shift settings.
Authors: Xihe Gu, Greg Fields, Yaman Jandali, Tara Javidi, Farinaz Koushanfar
Abstract: Trojan attacks are sophisticated training-time attacks on neural networks that embed backdoor triggers which force the network to produce a specific output on any input which includes the trigger. With the increasing relevance of deep networks which are too large to train with personal resources and which are trained on data too large to thoroughly audit, these training-time attacks pose a significant risk. In this work, we connect trojan attacks to Neural Collapse, a phenomenon wherein the final feature representations of over-parameterized neural networks converge to a simple geometric structure. We provide experimental evidence that trojan attacks disrupt this convergence for a variety of datasets and architectures. We then use this disruption to design a lightweight, broadly generalizable mechanism for cleansing trojan attacks from a wide variety of different network architectures and experimentally demonstrate its efficacy.
Authors: David Brandfonbrener, Nikhil Anand, Nikhil Vyas, Eran Malach, Sham Kakade
Abstract: While scaling laws provide a reliable methodology for predicting train loss across compute scales for a single data distribution, less is known about how these predictions should change as we change the distribution. In this paper, we derive a strategy for predicting one loss from another and apply it to predict across different pre-training datasets and from pre-training data to downstream task data. Our predictions extrapolate well even at 20x the largest FLOP budget used to fit the curves. More precisely, we find that there are simple shifted power law relationships between (1) the train losses of two models trained on two separate datasets when the models are paired by training compute (train-to-train), (2) the train loss and the test loss on any downstream distribution for a single model (train-to-test), and (3) the test losses of two models trained on two separate train datasets (test-to-test). The results hold up for pre-training datasets that differ substantially (some are entirely code and others have no code at all) and across a variety of downstream tasks. Finally, we find that in some settings these shifted power law relationships can yield more accurate predictions than extrapolating single-dataset scaling laws.
Authors: Dimple Vijay Kochar, Hanrui Wang, Anantha Chandrakasan, Xin Zhang
Abstract: Traditional approaches for designing analog circuits are time-consuming and require significant human expertise. Existing automation efforts using methods like Bayesian Optimization (BO) and Reinforcement Learning (RL) are sub-optimal and costly to generalize across different topologies and technology nodes. In our work, we introduce a novel approach, LEDRO, utilizing Large Language Models (LLMs) in conjunction with optimization techniques to iteratively refine the design space for analog circuit sizing. LEDRO is highly generalizable compared to other RL and BO baselines, eliminating the need for design annotation or model training for different topologies or technology nodes. We conduct a comprehensive evaluation of our proposed framework and baseline on 22 different Op-Amp topologies across four FinFET technology nodes. Results demonstrate the superior performance of LEDRO as it outperforms our best baseline by an average of 13% FoM improvement with 2.15x speed-up on low complexity Op-Amps and 48% FoM improvement with 1.7x speed-up on high complexity Op-Amps. This highlights LEDRO's effective performance, efficiency, and generalizability.
Authors: Edward McDugald, Arvind Mohan, Darren Engwirda, Agnese Marcato, Javier Santos
Abstract: We investigate the use of the Senseiver, a transformer neural network designed for sparse sensing applications, to estimate full-field surface height measurements of tsunami waves from sparse observations. The model is trained on a large ensemble of simulated data generated via a shallow water equations solver, which we show to be a faithful reproduction for the underlying dynamics by comparison to historical events. We train the model on a dataset consisting of 8 tsunami simulations whose epicenters correspond to historical USGS earthquake records, and where the model inputs are restricted to measurements obtained at actively deployed buoy locations. We test the Senseiver on a dataset consisting of 8 simulations not included in training, demonstrating its capability for extrapolation. The results show remarkable resolution of fine scale phase and amplitude features from the true field, provided that at least a few of the sensors have obtained a non-zero signal. Throughout, we discuss which forecasting techniques can be improved by this method, and suggest ways in which the flexibility of the architecture can be leveraged to incorporate arbitrary remote sensing data (eg. HF Radar and satellite measurements) as well as investigate optimal sensor placements.
Authors: Yuan Yuan, Jingtao Ding, Chonghua Han, Depeng Jin, Yong Li
Abstract: Urban spatio-temporal flow prediction, encompassing traffic flows and crowd flows, is crucial for optimizing city infrastructure and managing traffic and emergency responses. Traditional approaches have relied on separate models tailored to either grid-based data, representing cities as uniform cells, or graph-based data, modeling cities as networks of nodes and edges. In this paper, we build UniFlow, a foundational model for general urban flow prediction that unifies both grid-based and graphbased data. We first design a multi-view spatio-temporal patching mechanism to standardize different data into a consistent sequential format and then introduce a spatio-temporal transformer architecture to capture complex correlations and dynamics. To leverage shared spatio-temporal patterns across different data types and facilitate effective cross-learning, we propose SpatioTemporal Memory Retrieval Augmentation (ST-MRA). By creating structured memory modules to store shared spatio-temporal patterns, ST-MRA enhances predictions through adaptive memory retrieval. Extensive experiments demonstrate that UniFlow outperforms existing models in both grid-based and graph-based flow prediction, excelling particularly in scenarios with limited data availability, showcasing its superior performance and broad applicability. The datasets and code implementation have been released on https://github.com/YuanYuan98/UniFlow.
Authors: Runlong Yu, Chonghao Qiu, Robert Ladwig, Paul C. Hanson, Yiqun Xie, Yanhua Li, Xiaowei Jia
Abstract: This paper introduces a \textit{Process-Guided Learning (Pril)} framework that integrates physical models with recurrent neural networks (RNNs) to enhance the prediction of dissolved oxygen (DO) concentrations in lakes, which is crucial for sustaining water quality and ecosystem health. Unlike traditional RNNs, which may deliver high accuracy but often lack physical consistency and broad applicability, the \textit{Pril} method incorporates differential DO equations for each lake layer, modeling it as a first-order linear solution using a forward Euler scheme with a daily timestep. However, this method is sensitive to numerical instabilities. When drastic fluctuations occur, the numerical integration is neither mass-conservative nor stable. Especially during stratified conditions, exogenous fluxes into each layer cause significant within-day changes in DO concentrations. To address this challenge, we further propose an \textit{Adaptive Process-Guided Learning (April)} model, which dynamically adjusts timesteps from daily to sub-daily intervals with the aim of mitigating the discrepancies caused by variations in entrainment fluxes. \textit{April} uses a generator-discriminator architecture to identify days with significant DO fluctuations and employs a multi-step Euler scheme with sub-daily timesteps to effectively manage these variations. We have tested our methods on a wide range of lakes in the Midwestern USA, and demonstrated robust capability in predicting DO concentrations even with limited training data. While primarily focused on aquatic ecosystems, this approach is broadly applicable to diverse scientific and engineering disciplines that utilize process-based models, such as power engineering, climate science, and biomedicine.
Authors: Yuxuan Chen, Rongpeng Li, Zhifeng Zhao, Honggang Zhang
Abstract: We present MERLOT, a scalable mixture-of-expert (MoE) based refinement of distilled large language model optimized for encrypted traffic classification. By applying model distillation techniques in a teacher-student paradigm, compact models derived from GPT-2-base retain high classification accuracy while minimizing computational costs. These models function as specialized experts in an MoE architecture, dynamically assigned via a gating network. Unlike generation-based methods, our approach directly classifies encrypted traffic using the final decoder token with contextual feature embedding as input. Experiments on 10 datasets show superior or competitive performance over the state-of-the-art models while significantly reducing resource demands, underscoring its effectiveness and robustness.
Authors: Anique Tahir, Lu Cheng, Manuel Sandoval, Yasin N. Silva, Deborah L. Hall, Huan Liu
Abstract: Social media discourse involves people from different backgrounds, beliefs, and motives. Thus, often such discourse can devolve into toxic interactions. Generative Models, such as Llama and ChatGPT, have recently exploded in popularity due to their capabilities in zero-shot question-answering. Because these models are increasingly being used to ask questions of social significance, a crucial research question is whether they can understand social media dynamics. This work provides a critical analysis regarding generative LLM's ability to understand language and dynamics in social contexts, particularly considering cyberbullying and anti-cyberbullying (posts aimed at reducing cyberbullying) interactions. Specifically, we compare and contrast the capabilities of different large language models (LLMs) to understand three key aspects of social dynamics: language, directionality, and the occurrence of bullying/anti-bullying messages. We found that while fine-tuned LLMs exhibit promising results in some social media understanding tasks (understanding directionality), they presented mixed results in others (proper paraphrasing and bullying/anti-bullying detection). We also found that fine-tuning and prompt engineering mechanisms can have positive effects in some tasks. We believe that a understanding of LLM's capabilities is crucial to design future models that can be effectively used in social applications.
Authors: Zhuohan Gu, Jiayi Yao, Kuntai Du, Junchen Jiang
Abstract: As large language models (LLMs) show impressive performance on complex tasks, they still struggle with longer contextual understanding and high computational costs. To balance efficiency and quality, we introduce LLMSteer, a fine-tuning-free framework that enhances LLMs through query-independent attention steering. Tested on popular LLMs and datasets, LLMSteer narrows the performance gap with baselines by 65.9% and reduces the runtime delay by up to 4.8x compared to recent attention steering methods.
Authors: Allen Hao Huang
Abstract: Activation functions play a crucial role in introducing non-linearities to deep neural networks. We propose a novel approach to designing activation functions by focusing on their gradients and deriving the corresponding functions through integration. Our work introduces the Expanded Integral of the Exponential Linear Unit (xIELU), a trainable piecewise activation function derived by integrating trainable affine transformations applied on the ELU activation function. xIELU combines two key gradient properties: a trainable and linearly increasing gradient for positive inputs, similar to ReLU$^2$, and a trainable negative gradient flow for negative inputs, akin to xSiLU. Conceptually, xIELU can be viewed as extending ReLU$^2$ to effectively handle negative inputs. In experiments with 1.1B parameter Llama models trained on 126B tokens of FineWeb Edu, xIELU achieves lower perplexity compared to both ReLU$^2$ and SwiGLU when matched for the same compute cost and parameter count.
Authors: Xiang Li, Gagan Agrawal, Ruoming Jin, Rajiv Ramnath
Abstract: We consider the problem of constructing embeddings of large attributed graphs and supporting multiple downstream learning tasks. We develop a graph embedding method, which is based on extending deep metric and unbiased contrastive learning techniques to 1) work with attributed graphs, 2) enabling a mini-batch based approach, and 3) achieving scalability. Based on a multi-class tuplet loss function, we present two algorithms -- DMT for semi-supervised learning and DMAT-i for the unsupervised case. Analyzing our methods, we provide a generalization bound for the downstream node classification task and for the first time relate tuplet loss to contrastive learning. Through extensive experiments, we show high scalability of representation construction, and in applying the method for three downstream tasks (node clustering, node classification, and link prediction) better consistency over any single existing method.
Authors: Hamed Shirzad, Honghao Lin, Ameya Velingker, Balaji Venkatachalam, David Woodruff, Danica Sutherland
Abstract: Transductive tasks on graphs differ fundamentally from typical supervised machine learning tasks, as the independent and identically distributed (i.i.d.) assumption does not hold among samples. Instead, all train/test/validation samples are present during training, making them more akin to a semi-supervised task. These differences make the analysis of the models substantially different from other models. Recently, Graph Transformers have significantly improved results on these datasets by overcoming long-range dependency problems. However, the quadratic complexity of full Transformers has driven the community to explore more efficient variants, such as those with sparser attention patterns. While the attention matrix has been extensively discussed, the hidden dimension or width of the network has received less attention. In this work, we establish some theoretical bounds on how and under what conditions the hidden dimension of these networks can be compressed. Our results apply to both sparse and dense variants of Graph Transformers.
Authors: Lee Cohen, Yishay Mansour, Shay Moran, Han Shao
Abstract: Precision and Recall are foundational metrics in machine learning where both accurate predictions and comprehensive coverage are essential, such as in recommender systems and multi-label learning. In these tasks, balancing precision (the proportion of relevant items among those predicted) and recall (the proportion of relevant items successfully predicted) is crucial. A key challenge is that one-sided feedback--where only positive examples are observed during training--is inherent in many practical problems. For instance, in recommender systems like YouTube, training data only consists of videos that a user has actively selected, while unselected items remain unseen. Despite this lack of negative feedback in training, avoiding undesirable recommendations at test time is essential. We introduce a PAC learning framework where each hypothesis is represented by a graph, with edges indicating positive interactions, such as between users and items. This framework subsumes the classical binary and multi-class PAC learning models as well as multi-label learning with partial feedback, where only a single random correct label per example is observed, rather than all correct labels. Our work uncovers a rich statistical and algorithmic landscape, with nuanced boundaries on what can and cannot be learned. Notably, classical methods like Empirical Risk Minimization fail in this setting, even for simple hypothesis classes with only two hypotheses. To address these challenges, we develop novel algorithms that learn exclusively from positive data, effectively minimizing both precision and recall losses. Specifically, in the realizable setting, we design algorithms that achieve optimal sample complexity guarantees. In the agnostic case, we show that it is impossible to achieve additive error guarantees--as is standard in PAC learning--and instead obtain meaningful multiplicative approximations.
Authors: Jared Fernandez, Luca Wehrstedt, Leonid Shamis, Mostafa Elhoushi, Kalyan Saladi, Yonatan Bisk, Emma Strubell, Jacob Kahn
Abstract: Dramatic increases in the capabilities of neural network models in recent years are driven by scaling model size, training data, and corresponding computational resources. To develop the exceedingly large networks required in modern applications, such as large language models (LLMs), model training is distributed across tens of thousands of hardware accelerators (e.g. GPUs), requiring orchestration of computation and communication across large computing clusters. In this work, we demonstrate that careful consideration of hardware configuration and parallelization strategy is critical for effective (i.e. compute- and cost-efficient) scaling of model size, training data, and total computation. We conduct an extensive empirical study of the performance of large-scale LLM training workloads across model size, hardware configurations, and distributed parallelization strategies. We demonstrate that: (1) beyond certain scales, overhead incurred from certain distributed communication strategies leads parallelization strategies previously thought to be sub-optimal in fact become preferable; and (2) scaling the total number of accelerators for large model training quickly yields diminishing returns even when hardware and parallelization strategies are properly optimized, implying poor marginal performance per additional unit of power or GPU-hour.
Authors: Shuman Peng, Arash Khoeini, Sharan Vaswani, Martin Ester
Abstract: The quality of self-supervised pre-trained embeddings on out-of-distribution (OOD) data is poor without fine-tuning. A straightforward and simple approach to improving the generalization of pre-trained representation to OOD data is the use of deep ensembles. However, obtaining an effective ensemble in the embedding space with only unlabeled data remains an unsolved problem. We first perform a theoretical analysis that reveals the relationship between individual hyperspherical embedding spaces in an ensemble. We then design a principled method to align these embedding spaces in an unsupervised manner. Experimental results on the MNIST dataset show that our embedding-space ensemble method improves pre-trained embedding quality on in-distribution and OOD data compared to single encoders.
Authors: Lunjia Hu, Kevin Tian, Chutong Yang
Abstract: Recent work on supervised learning [GKR+22] defined the notion of omnipredictors, i.e., predictor functions $p$ over features that are simultaneously competitive for minimizing a family of loss functions $\mathcal{L}$ against a comparator class $\mathcal{C}$. Omniprediction requires approximating the Bayes-optimal predictor beyond the loss minimization paradigm, and has generated significant interest in the learning theory community. However, even for basic settings such as agnostically learning single-index models (SIMs), existing omnipredictor constructions require impractically-large sample complexities and runtimes, and output complex, highly-improper hypotheses. Our main contribution is a new, simple construction of omnipredictors for SIMs. We give a learner outputting an omnipredictor that is $\varepsilon$-competitive on any matching loss induced by a monotone, Lipschitz link function, when the comparator class is bounded linear predictors. Our algorithm requires $\approx \varepsilon^{-4}$ samples and runs in nearly-linear time, and its sample complexity improves to $\approx \varepsilon^{-2}$ if link functions are bi-Lipschitz. This significantly improves upon the only prior known construction, due to [HJKRR18, GHK+23], which used $\gtrsim \varepsilon^{-10}$ samples. We achieve our construction via a new, sharp analysis of the classical Isotron algorithm [KS09, KKKS11] in the challenging agnostic learning setting, of potential independent interest. Previously, Isotron was known to properly learn SIMs in the realizable setting, as well as constant-factor competitive hypotheses under the squared loss [ZWDD24]. As they are based on Isotron, our omnipredictors are multi-index models with $\approx \varepsilon^{-2}$ prediction heads, bringing us closer to the tantalizing goal of proper omniprediction for general loss families and comparators.
Authors: Ziqi Jia, Xiaoyang Qu, Chenghao Liu, Jianzong Wang
Abstract: Label Distribution Learning (LDL) is an effective approach for handling label ambiguity, as it can analyze all labels at once and indicate the extent to which each label describes a given sample. Most existing LDL methods consider the number of labels to be static. However, in various LDL-specific contexts (e.g., disease diagnosis), the label count grows over time (such as the discovery of new diseases), a factor that existing methods overlook. Learning samples with new labels directly means learning all labels at once, thus wasting more time on the old labels and even risking overfitting the old labels. At the same time, learning new labels by the LDL model means reconstructing the inter-label relationships. How to make use of constructed relationships is also a crucial challenge. To tackle these challenges, we introduce Incremental Label Distribution Learning (ILDL), analyze its key issues regarding training samples and inter-label relationships, and propose Scalable Graph Label Distribution Learning (SGLDL) as a practical framework for implementing ILDL. Specifically, in SGLDL, we develop a New-label-aware Gradient Compensation Loss to speed up the learning of new labels and represent inter-label relationships as a graph to reduce the time required to reconstruct inter-label relationships. Experimental results on the classical LDL dataset show the clear advantages of unique algorithms and illustrate the importance of a dedicated design for the ILDL problem.
Authors: Zheng Zhang, Qiong Wu, Pingyi Fan, Nan Cheng, Wen Chen, Khaled B. Letaief
Abstract: To address communication latency issues, the Third Generation Partnership Project (3GPP) has defined Cellular-Vehicle to Everything (C-V2X) technology, which includes Vehicle-to-Vehicle (V2V) communication for direct vehicle-to-vehicle communication. However, this method requires vehicles to autonomously select communication resources based on the Semi-Persistent Scheduling (SPS) protocol, which may lead to collisions due to different vehicles sharing the same communication resources, thereby affecting communication effectiveness. Non-Orthogonal Multiple Access (NOMA) is considered a potential solution for handling large-scale vehicle communication, as it can enhance the Signal-to-Interference-plus-Noise Ratio (SINR) by employing Successive Interference Cancellation (SIC), thereby reducing the negative impact of communication collisions. When evaluating vehicle communication performance, traditional metrics such as reliability and transmission delay present certain contradictions. Introducing the new metric Age of Information (AoI) provides a more comprehensive evaluation of communication system. Additionally, to ensure service quality, user terminals need to possess high computational capabilities, which may lead to increased energy consumption, necessitating a trade-off between communication energy consumption and effectiveness. Given the complexity and dynamics of communication systems, Deep Reinforcement Learning (DRL) serves as an intelligent learning method capable of learning optimal strategies in dynamic environments. Therefore, this paper analyzes the effects of multi-priority queues and NOMA on AoI in the C-V2X vehicular communication system and proposes an energy consumption and AoI optimization method based on DRL. Finally, through comparative simulations with baseline methods, the proposed approach demonstrates its advances in terms of energy consumption and AoI.
Authors: Zhi Luo, Xiyuan Yang, Pan Zhou, Di Wang
Abstract: Manipulating the interaction trajectories between the intelligent agent and the environment can control the agent's training and behavior, exposing the potential vulnerabilities of reinforcement learning (RL). For example, in Cyber-Physical Systems (CPS) controlled by RL, the attacker can manipulate the actions of the adopted RL to other actions during the training phase, which will lead to bad consequences. Existing work has studied action-manipulation attacks in tabular settings, where the states and actions are discrete. As seen in many up-and-coming RL applications, such as autonomous driving, continuous action space is widely accepted, however, its action-manipulation attacks have not been thoroughly investigated yet. In this paper, we consider this crucial problem in both white-box and black-box scenarios. Specifically, utilizing the knowledge derived exclusively from trajectories, we propose a black-box attack algorithm named LCBT, which uses the Monte Carlo tree search method for efficient action searching and manipulation. Additionally, we demonstrate that for an agent whose dynamic regret is sub-linearly related to the total number of steps, LCBT can teach the agent to converge to target policies with only sublinear attack cost, i.e., $O\left(\mathcal{R}(T) + MH^3K^E\log (MT)\right)(0
Authors: Charles O'Neill, David Klindt
Abstract: A recent line of work has shown promise in using sparse autoencoders (SAEs) to uncover interpretable features in neural network representations. However, the simple linear-nonlinear encoding mechanism in SAEs limits their ability to perform accurate sparse inference. In this paper, we investigate sparse inference and learning in SAEs through the lens of sparse coding. Specifically, we show that SAEs perform amortised sparse inference with a computationally restricted encoder and, using compressed sensing theory, we prove that this mapping is inherently insufficient for accurate sparse inference, even in solvable cases. Building on this theory, we empirically explore conditions where more sophisticated sparse inference methods outperform traditional SAE encoders. Our key contribution is the decoupling of the encoding and decoding processes, which allows for a comparison of various sparse encoding strategies. We evaluate these strategies on two dimensions: alignment with true underlying sparse features and correct inference of sparse codes, while also accounting for computational costs during training and inference. Our results reveal that substantial performance gains can be achieved with minimal increases in compute cost. We demonstrate that this generalises to SAEs applied to large language models (LLMs), where advanced encoders achieve similar interpretability. This work opens new avenues for understanding neural network representations and offers important implications for improving the tools we use to analyse the activations of large language models.
Authors: Zepeng Zhang, Olga Fink
Abstract: Over the last decade, graph neural networks (GNNs) have made significant progress in numerous graph machine learning tasks. In real-world applications, where domain shifts occur and labels are often unavailable for a new target domain, graph domain adaptation (GDA) approaches have been proposed to facilitate knowledge transfer from the source domain to the target domain. Previous efforts in tackling distribution shifts across domains have mainly focused on aligning the node embedding distributions generated by the GNNs in the source and target domains. However, as the core part of GDA approaches, the impact of the underlying GNN architecture has received limited attention. In this work, we explore this orthogonal direction, i.e., how to facilitate GDA with architectural enhancement. In particular, we consider a class of GNNs that are designed explicitly based on optimization problems, namely unfolded GNNs (UGNNs), whose training process can be represented as bi-level optimization. Empirical and theoretical analyses demonstrate that when transferring from the source domain to the target domain, the lower-level objective value generated by the UGNNs significantly increases, resulting in an increase in the upper-level objective as well. Motivated by this observation, we propose a simple yet effective strategy called cascaded propagation (CP), which is guaranteed to decrease the lower-level objective value. The CP strategy is widely applicable to general UGNNs, and we evaluate its efficacy with three representative UGNN architectures. Extensive experiments on five real-world datasets demonstrate that the UGNNs integrated with CP outperform state-of-the-art GDA baselines.
Authors: Kai Tanaka, Mineichi Kudo, Keigo Kimura, Atsuyoshi Nakamura
Abstract: The proportion of elderly people is increasing worldwide, particularly those living alone in Japan. As elderly people get older, their risks of physical disabilities and health issues increase. To automatically discover these issues at a low cost in daily life, sensor-based detection in a smart home is promising. As part of the effort towards early detection of abnormal behaviors, we propose a simulator-based detection systems for six typical anomalies: being semi-bedridden, being housebound, forgetting, wandering, fall while walking and fall while standing. Our detection system can be customized for various room layout, sensor arrangement and resident's characteristics by training detection classifiers using the simulator with the parameters fitted to individual cases. Considering that the six anomalies that our system detects have various occurrence durations, such as being housebound for weeks or lying still for seconds after a fall, the detection classifiers of our system produce anomaly labels depending on each anomaly's occurrence duration, e.g., housebound per day and falls per second. We propose a method that standardizes the processing of sensor data, and uses a simple detection approach. Although the validity depends on the realism of the simulation, numerical evaluations using sensor data that includes a variety of resident behavior patterns over nine years as test data show that (1) the methods for detecting wandering and falls are comparable to previous methods, and (2) the methods for detecting being semi-bedridden, being housebound, and forgetting achieve a sensitivity of over 0.9 with fewer than one false alarm every 50 days.
Authors: Nabeel Seedat, Caterina Tozzi, Andrea Hita Ardiaca, Mihaela van der Schaar, James Weatherall, Adam Taylor
Abstract: The reuse of historical clinical trial data has significant potential to accelerate medical research and drug development. However, interoperability challenges, particularly with missing medical codes, hinders effective data integration across studies. While Large Language Models (LLMs) offer a promising solution for automated coding without labeled data, current approaches face challenges on complex coding tasks. We introduce ALIGN, a novel compositional LLM-based system for automated, zero-shot medical coding. ALIGN follows a three-step process: (1) diverse candidate code generation; (2) self-evaluation of codes and (3) confidence scoring and uncertainty estimation enabling human deferral to ensure reliability. We evaluate ALIGN on harmonizing medication terms into Anatomical Therapeutic Chemical (ATC) and medical history terms into Medical Dictionary for Regulatory Activities (MedDRA) codes extracted from 22 immunology trials. ALIGN outperformed the LLM baselines, while also providing capabilities for trustworthy deployment. For MedDRA coding, ALIGN achieved high accuracy across all levels, matching RAG and excelling at the most specific levels (87-90% for HLGT). For ATC coding, ALIGN demonstrated superior performance, particularly at lower hierarchy levels (ATC Level 4), with 72-73% overall accuracy and 86-89% accuracy for common medications, outperforming baselines by 7-22%. ALIGN's uncertainty-based deferral improved accuracy by 17% to 90% accuracy with 30% deferral, notably enhancing performance on uncommon medications. ALIGN achieves this cost-efficiently at \$0.0007 and \$0.02 per code for GPT-4o-mini and GPT-4o, reducing barriers to clinical adoption. ALIGN advances automated medical coding for clinical trial data, contributing to enhanced data interoperability and reusability, positioning it as a promising tool to improve clinical research and accelerate drug development.
Authors: Peng Wang, Li Shen, Zerui Tao, Yan Sun, Guodong Zheng, Dacheng Tao
Abstract: Averaging iterations of Stochastic Gradient Descent (SGD) have achieved empirical success in training deep learning models, such as Stochastic Weight Averaging (SWA), Exponential Moving Average (EMA), and LAtest Weight Averaging (LAWA). Especially, with a finite weight averaging method, LAWA can attain faster convergence and better generalization. However, its theoretical explanation is still less explored since there are fundamental differences between finite and infinite settings. In this work, we first generalize SGD and LAWA as Finite Weight Averaging (FWA) and explain their advantages compared to SGD from the perspective of optimization and generalization. A key challenge is the inapplicability of traditional methods in the sense of expectation or optimal values for infinite-dimensional settings in analyzing FWA's convergence. Second, the cumulative gradients introduced by FWA introduce additional confusion to the generalization analysis, especially making it more difficult to discuss them under different assumptions. Extending the final iteration convergence analysis to the FWA, this paper, under a convexity assumption, establishes a convergence bound $\mathcal{O}(\log\left(\frac{T}{k}\right)/\sqrt{T})$, where $k\in[1, T/2]$ is a constant representing the last $k$ iterations. Compared to SGD with $\mathcal{O}(\log(T)/\sqrt{T})$, we prove theoretically that FWA has a faster convergence rate and explain the effect of the number of average points. In the generalization analysis, we find a recursive representation for bounding the cumulative gradient using mathematical induction. We provide bounds for constant and decay learning rates and the convex and non-convex cases to show the good generalization performance of FWA. Finally, experimental results on several benchmarks verify our theoretical results.
Authors: Erica Coppolillo, Marco Minici, Federico Cinus, Francesco Bonchi, Giuseppe Manco
Abstract: Large Language Models (LLMs) exhibit significant persuasion capabilities in one-on-one interactions, but their influence within social networks remains underexplored. This study investigates the potential social impact of LLMs in these environments, where interconnected users and complex opinion dynamics pose unique challenges. In particular, we address the following research question: can LLMs learn to generate meaningful content that maximizes user engagement on social networks? To answer this question, we define a pipeline to guide the LLM-based content generation which employs reinforcement learning with simulated feedback. In our framework, the reward is based on an engagement model borrowed from the literature on opinion dynamics and information propagation. Moreover, we force the text generated by the LLM to be aligned with a given topic and to satisfy a minimum fluency requirement. Using our framework, we analyze the capabilities and limitations of LLMs in tackling the given task, specifically considering the relative positions of the LLM as an agent within the social network and the distribution of opinions in the network on the given topic. Our findings show the full potential of LLMs in creating social engagement. Notable properties of our approach are that the learning procedure is adaptive to the opinion distribution of the underlying network and agnostic to the specifics of the engagement model, which is embedded as a plug-and-play component. In this regard, our approach can be easily refined for more complex engagement tasks and interventions in computational social science. The code used for the experiments is publicly available at https://anonymous.4open.science/r/EDCG/.
Authors: Riya Mahesh, Rahul Vashisht, Chandrashekar Lakshminarayanan
Abstract: Temporal causal analysis means understanding the underlying causes behind observed variables over time. Deep learning based methods such as transformers are increasingly used to capture temporal dynamics and causal relationships beyond mere correlations. Recent works suggest self-attention weights of transformers as a useful indicator of causal links. We leverage this to propose a novel modification to the self-attention module to establish causal links between the variables of multivariate time-series data with varying lag dependencies. Our Sparse Attention Transformer captures causal relationships using a two-fold approach - performing temporal attention first followed by attention between the variables across the time steps masking them individually to compute Granger Causality indices. The key novelty in our approach is the ability of the model to assert importance and pick the most significant past time instances for its prediction task against manually feeding a fixed time lag value. We demonstrate the effectiveness of our approach via extensive experimentation on several synthetic benchmark datasets. Furthermore, we compare the performance of our model with the traditional Vector Autoregression based Granger Causality method that assumes fixed lag length.
Authors: \`Alex Pujol Vidal, Anders S. Johansen, Mohammad N. S. Jahromi, Sergio Escalera, Kamal Nasrollahi, Thomas B. Moeslund
Abstract: We investigate the effectiveness of Explainable AI (XAI) in verifying Machine Unlearning (MU) within the context of harbor front monitoring, focusing on data privacy and regulatory compliance. With the increasing need to adhere to privacy legislation such as the General Data Protection Regulation (GDPR), traditional methods of retraining ML models for data deletions prove impractical due to their complexity and resource demands. MU offers a solution by enabling models to selectively forget specific learned patterns without full retraining. We explore various removal techniques, including data relabeling, and model perturbation. Then, we leverage attribution-based XAI to discuss the effects of unlearning on model performance. Our proof-of-concept introduces feature importance as an innovative verification step for MU, expanding beyond traditional metrics and demonstrating techniques' ability to reduce reliance on undesired patterns. Additionally, we propose two novel XAI-based metrics, Heatmap Coverage (HC) and Attention Shift (AS), to evaluate the effectiveness of these methods. This approach not only highlights how XAI can complement MU by providing effective verification, but also sets the stage for future research to enhance their joint integration.
Authors: Mai Elkady, Thu Bui, Bruno Ribeiro, David I. Inouye
Abstract: There has been a growing excitement that implicit graph generative models could be used to design or discover new molecules for medicine or material design. Because these molecules have not been discovered, they naturally lie in unexplored or scarcely supported regions of the distribution of known molecules. However, prior evaluation methods for implicit graph generative models have focused on validating statistics computed from the thick support (e.g., mean and variance of a graph property). Therefore, there is a mismatch between the goal of generating novel graphs and the evaluation methods. To address this evaluation gap, we design a novel evaluation method called Vertical Validation (VV) that systematically creates thin support regions during the train-test splitting procedure and then reweights generated samples so that they can be compared to the held-out test data. This procedure can be seen as a generalization of the standard train-test procedure except that the splits are dependent on sample features. We demonstrate that our method can be used to perform model selection if performance on thin support regions is the desired goal. As a side benefit, we also show that our approach can better detect overfitting as exemplified by memorization.
Authors: Sasa Ilic, Abdulkerim Karaman, Johannes P\"oppelbaum, Jan Niclas Reimann, Michael Marr\'e, Andreas Schwung
Abstract: This study presents a novel approach for predicting wall thickness changes in tubes during the nosing process. Specifically, we first provide a thorough analysis of nosing processes and the influencing parameters. We further set-up a Finite Element Method (FEM) simulation to better analyse the effects of varying process parameters. As however traditional FEM simulations, while accurate, are time-consuming and computationally intensive, which renders them inapplicable for real-time application, we present a novel modeling framework based on specifically designed graph neural networks as surrogate models. To this end, we extend the neural network architecture by directly incorporating information about the nosing process by adding different types of edges and their corresponding encoders to model object interactions. This augmentation enhances model accuracy and opens the possibility for employing precise surrogate models within closed-loop production processes. The proposed approach is evaluated using a new evaluation metric termed area between thickness curves (ABTC). The results demonstrate promising performance and highlight the potential of neural networks as surrogate models in predicting wall thickness changes during nosing forging processes.
Authors: Ricardo Monta\~nana, Jos\'e A. G\'amez, Jos\'e M. Puerta
Abstract: We propose ODTE, a new ensemble that uses oblique decision trees as base classifiers. Additionally, we introduce STree, the base algorithm for growing oblique decision trees, which leverages support vector machines to define hyperplanes within the decision nodes. We embed a multiclass strategy -- one-vs-one or one-vs-rest -- at the decision nodes, allowing the model to directly handle non-binary classification tasks without the need to cluster instances into two groups, as is common in other approaches from the literature. In each decision node, only the best-performing model SVM -- the one that minimizes an impurity measure for the n-ary classification -- is retained, even if the learned SVM addresses a binary classification subtask. An extensive experimental study involving 49 datasets and various state-of-the-art algorithms for oblique decision tree ensembles has been conducted. Our results show that ODTE ranks consistently above its competitors, achieving significant performance gains when hyperparameters are carefully tuned. Moreover, the oblique decision trees learned through STree are more compact than those produced by other algorithms evaluated in our experiments.
Authors: Alireza Rashidi Laleh, Majid Nili Ahmadabadi
Abstract: Reinforcement learning (RL) is one of the active fields in machine learning, demonstrating remarkable potential in tackling real-world challenges. Despite its promising prospects, this methodology has encountered with issues and challenges, hindering it from achieving the best performance. In particular, these approaches lack decent performance when navigating environments and solving tasks with large observation space, often resulting in sample-inefficiency and prolonged learning times. This issue, commonly referred to as the curse of dimensionality, complicates decision-making for RL agents, necessitating a careful balance between attention and decision-making. RL agents, when augmented with human or large language models' (LLMs) feedback, may exhibit resilience and adaptability, leading to enhanced performance and accelerated learning. Such feedback, conveyed through various modalities or granularities including natural language, serves as a guide for RL agents, aiding them in discerning relevant environmental cues and optimizing decision-making processes. In this survey paper, we mainly focus on problems of two-folds: firstly, we focus on humans or an LLMs assistance, investigating the ways in which these entities may collaborate with the RL agent in order to foster optimal behavior and expedite learning; secondly, we delve into the research papers dedicated to addressing the intricacies of environments characterized by large observation space.
Authors: Hojjat Karami, David Atienza, Anisoara Ionescu
Abstract: Generating synthetic Electronic Health Records (EHRs) offers significant potential for data augmentation, privacy-preserving data sharing, and improving machine learning model training. We propose a novel tokenization strategy tailored for structured EHR data, which encompasses diverse data types such as covariates, ICD codes, and irregularly sampled time series. Using a GPT-like decoder-only transformer model, we demonstrate the generation of high-quality synthetic EHRs. Our approach is evaluated using the MIMIC-III dataset, and we benchmark the fidelity, utility, and privacy of the generated data against state-of-the-art models.
Authors: Farrukh A. Chishtie, Dominique Brunet, Rachel H. White, Daniel Michelson, Jing Jiang, Vicky Lucas, Emily Ruboonga, Sayana Imaash, Melissa Westland, Timothy Chui, Rana Usman Ali, Mujtaba Hassan, Roland Stull, David Hudak
Abstract: Heatwaves, prolonged periods of extreme heat, have intensified in frequency and severity due to climate change, posing substantial risks to public health, ecosystems, and infrastructure. Despite advancements in Machine Learning (ML) modeling, accurate heatwave forecasting at weather scales (1--15 days) remains challenging due to the non-linear interactions between atmospheric drivers and the rarity of these extreme events. Traditional models relying on heuristic feature engineering often fail to generalize across diverse climates and capture the complexities of heatwave dynamics. This study introduces the Distribution-Informed Graph Neural Network (DI-GNN), a novel framework that integrates principles from Extreme Value Theory (EVT) into the graph neural network architecture. DI-GNN incorporates Generalized Pareto Distribution (GPD)-derived descriptors into the feature space, adjacency matrix, and loss function to enhance its sensitivity to rare heatwave occurrences. By prioritizing the tails of climatic distributions, DI-GNN addresses the limitations of existing methods, particularly in imbalanced datasets where traditional metrics like accuracy are misleading. Empirical evaluations using weather station data from British Columbia, Canada, demonstrate the superior performance of DI-GNN compared to baseline models. DI-GNN achieved significant improvements in balanced accuracy, recall, and precision, with high AUC and average precision scores, reflecting its robustness in distinguishing heatwave events.
Authors: Rodolfo Valiente, Praveen K. Pilly
Abstract: Metacognition--the awareness and regulation of one's cognitive processes--is central to human adaptability in unknown situations. In contrast, current autonomous agents often struggle in novel environments due to their limited capacity for adaptation. We hypothesize that metacognition is a critical missing ingredient in adaptive autonomous systems, equipping them with the cognitive flexibility needed to tackle unfamiliar challenges. Given the broad scope of metacognitive abilities, we focus on two key aspects: competence awareness and strategy selection for novel tasks. To this end, we propose the Metacognition for Unknown Situations and Environments (MUSE) framework, which integrates metacognitive processes--specifically self-awareness and self-regulation--into autonomous agents. We present two initial implementations of MUSE: one based on world modeling and another leveraging large language models (LLMs), both instantiating the metacognitive cycle. Our system continuously learns to assess its competence on a given task and uses this self-awareness to guide iterative cycles of strategy selection. MUSE agents show significant improvements in self-awareness and self-regulation, enabling them to solve novel, out-of-distribution tasks more effectively compared to Dreamer-v3-based reinforcement learning and purely prompt-based LLM agent approaches. This work highlights the promise of approaches inspired by cognitive and neural systems in enabling autonomous systems to adapt to new environments, overcoming the limitations of current methods that rely heavily on extensive training data.
Authors: Rushabh Solanki, Elliot Creager
Abstract: The deployment of AI in consumer products is currently focused on the use of so-called foundation models, large neural networks pre-trained on massive corpora of digital records. This emphasis on scaling up datasets and pre-training computation raises the risk of further consolidating the industry, and enabling monopolistic (or oligopolistic) behavior. Judges and regulators seeking to improve market competition may employ various remedies. This paper explores dissolution -- the breaking up of a monopolistic entity into smaller firms -- as one such remedy, focusing in particular on the technical challenges and opportunities involved in the breaking up of large models and datasets. We show how the framework of Conscious Data Contribution can enable user autonomy during under dissolution. Through a simulation study, we explore how fine-tuning and the phenomenon of "catastrophic forgetting" could actually prove beneficial as a type of machine unlearning that allows users to specify which data they want used for what purposes.
Authors: Yahui Bai, Yuhe Gao, Runzhe Wan, Sheng Zhang, Rui Song
Abstract: In recent years, there has been a growing trend of applying Reinforcement Learning (RL) in financial applications. This approach has shown great potential to solve decision-making tasks in finance. In this survey, we present a comprehensive study of the applications of RL in finance and conduct a series of meta-analyses to investigate the common themes in the literature, such as the factors that most significantly affect RL's performance compared to traditional methods. Moreover, we identify challenges including explainability, Markov Decision Process (MDP) modeling, and robustness that hinder the broader utilization of RL in the financial industry and discuss recent advancements in overcoming these challenges. Finally, we propose future research directions, such as benchmarking, contextual RL, multi-agent RL, and model-based RL to address these challenges and to further enhance the implementation of RL in finance.
Authors: Junhua Liu
Abstract: Financial AI empowers sophisticated approaches to financial market forecasting, portfolio optimization, and automated trading. This survey provides a systematic analysis of these developments across three primary dimensions: predictive models that capture complex market dynamics, decision-making frameworks that optimize trading and investment strategies, and knowledge augmentation systems that leverage unstructured financial information. We examine significant innovations including foundation models for financial time series, graph-based architectures for market relationship modeling, and hierarchical frameworks for portfolio optimization. Analysis reveals crucial trade-offs between model sophistication and practical constraints, particularly in high-frequency trading applications. We identify critical gaps and open challenges between theoretical advances and industrial implementation, outlining open challenges and opportunities for improving both model performance and practical applicability.
Authors: Mabsur Fatin Bin Hossain, Lubna Zahan Lamia, Md Mahmudur Rahman, Md Mosaddek Khan
Abstract: Time series forecasting is a key tool in financial markets, helping to predict asset prices and guide investment decisions. In highly volatile markets, such as cryptocurrencies like Bitcoin (BTC) and Ethereum (ETH), forecasting becomes more difficult due to extreme price fluctuations driven by market sentiment, technological changes, and regulatory shifts. Traditionally, forecasting relied on statistical methods, but as markets became more complex, deep learning models like LSTM, Bi-LSTM, and the newer FinBERT-LSTM emerged to capture intricate patterns. Building upon recent advancements and addressing the volatility inherent in cryptocurrency markets, we propose a hybrid model that combines Bidirectional Long Short-Term Memory (Bi-LSTM) networks with FinBERT to enhance forecasting accuracy for these assets. This approach fills a key gap in forecasting volatile financial markets by blending advanced time series models with sentiment analysis, offering valuable insights for investors and analysts navigating unpredictable markets.
Authors: Bartosz Bieganowski, Robert \'Slepaczuk
Abstract: This paper investigates the enhancement of financial time series forecasting with the use of neural networks through supervised autoencoders (SAE), to improve investment strategy performance. Using the Sharpe and Information Ratios, it specifically examines the impact of noise augmentation and triple barrier labeling on risk-adjusted returns. The study focuses on Bitcoin, Litecoin, and Ethereum as the traded assets from January 1, 2016, to April 30, 2022. Findings indicate that supervised autoencoders, with balanced noise augmentation and bottleneck size, significantly boost strategy effectiveness. However, excessive noise and large bottleneck sizes can impair performance.
Authors: Rishit Kapoor (Vellore Institute of Technology), Jesher Joshua (Vellore Institute of Technology), Muralidharan Vijayarangan (Vellore Institute of Technology), Natarajan B (Vellore Institute of Technology)
Abstract: This research work introduces a novel approach to the classification of Alzheimer's disease by using the advanced deep learning techniques combined with secure data processing methods. This research work primary uses transfer learning models such as ResNet, ImageNet, and VNet to extract high-level features from medical image data. Thereafter, these pre-trained models were fine-tuned for Alzheimer's related subtle patterns such that the model is capable of robust feature extraction over varying data sources. Further, the federated learning approaches were incorporated to tackle a few other challenges related to classification, aimed to provide better prediction performance and protect data privacy. The proposed model was built using federated learning without sharing sensitive patient data. This way, the decentralized model benefits from the large and diversified dataset that it is trained upon while ensuring confidentiality. The cipher-based encryption mechanism is added that allows us to secure the transportation of data and further ensure the privacy and integrity of patient information throughout training and classification. The results of the experiments not only help to improve the accuracy of the classification of Alzheimer's but at the same time provides a framework for secure and collaborative analysis of health care data.
Authors: Zeel B Patel, Yash Bachwana, Nitish Sharma, Sarath Guttikunda, Nipun Batra
Abstract: Nearly 6.7 million lives are lost due to air pollution every year. While policymakers are working on the mitigation strategies, public awareness can help reduce the exposure to air pollution. Air pollution data from government-installed sensors is often publicly available in raw format, but there is a non-trivial barrier for various stakeholders in deriving meaningful insights from that data. In this work, we present VayuBuddy, a Large Language Model (LLM)-powered chatbot system to reduce the barrier between the stakeholders and air quality sensor data. VayuBuddy receives the questions in natural language, analyses the structured sensory data with a LLM-generated Python code and provides answers in natural language. We use the data from Indian government air quality sensors. We benchmark the capabilities of 7 LLMs on 45 diverse question-answer pairs prepared by us. Additionally, VayuBuddy can also generate visual analysis such as line-plots, map plot, bar charts and many others from the sensory data as we demonstrate in this work.
Authors: Nay Myat Min, Long H. Pham, Yige Li, Jun Sun
Abstract: Recent studies reveal that Large Language Models (LLMs) are susceptible to backdoor attacks, where adversaries embed hidden triggers that manipulate model responses. Existing backdoor defense methods are primarily designed for vision or classification tasks, and are thus ineffective for text generation tasks, leaving LLMs vulnerable. We introduce Internal Consistency Regularization (CROW), a novel defense using consistency regularization finetuning to address layer-wise inconsistencies caused by backdoor triggers. CROW leverages the intuition that clean models exhibit smooth, consistent transitions in hidden representations across layers, whereas backdoored models show noticeable fluctuation when triggered. By enforcing internal consistency through adversarial perturbations and regularization, CROW neutralizes backdoor effects without requiring clean reference models or prior trigger knowledge, relying only on a small set of clean data. This makes it practical for deployment across various LLM architectures. Experimental results demonstrate that CROW consistently achieves a significant reductions in attack success rates across diverse backdoor strategies and tasks, including negative sentiment, targeted refusal, and code injection, on models such as Llama-2 (7B, 13B), CodeLlama (7B, 13B) and Mistral-7B, while preserving the model's generative capabilities.
Authors: Mahsa Nasri, Mehmet Kosa, Leanne Chukoskie, Mohsen Moghaddam, Casper Harteveld
Abstract: Virtual Reality (VR) has been a beneficial training tool in fields such as advanced manufacturing. However, users may experience a high cognitive load due to various factors, such as the use of VR hardware or tasks within the VR environment. Studies have shown that eye-tracking has the potential to detect cognitive load, but in the context of VR and complex spatiotemporal tasks (e.g., assembly and disassembly), it remains relatively unexplored. Here, we present an ongoing study to detect users' cognitive load using an eye-tracking-based machine learning approach. We developed a VR training system for cold spray and tested it with 22 participants, obtaining 19 valid eye-tracking datasets and NASA-TLX scores. We applied Multi-Layer Perceptron (MLP) and Random Forest (RF) models to compare the accuracy of predicting cognitive load (i.e., NASA-TLX) using pupil dilation and fixation duration. Our preliminary analysis demonstrates the feasibility of using eye tracking to detect cognitive load in complex spatiotemporal VR experiences and motivates further exploration.
Authors: Jeonghwan Lee, Cong Ma
Abstract: We consider estimation of a linear functional of the treatment effect using adaptively collected data. This task finds a variety of applications including the off-policy evaluation (\textsf{OPE}) in contextual bandits, and estimation of the average treatment effect (\textsf{ATE}) in causal inference. While a certain class of augmented inverse propensity weighting (\textsf{AIPW}) estimators enjoys desirable asymptotic properties including the semi-parametric efficiency, much less is known about their non-asymptotic theory with adaptively collected data. To fill in the gap, we first establish generic upper bounds on the mean-squared error of the class of AIPW estimators that crucially depends on a sequentially weighted error between the treatment effect and its estimates. Motivated by this, we also propose a general reduction scheme that allows one to produce a sequence of estimates for the treatment effect via online learning to minimize the sequentially weighted estimation error. To illustrate this, we provide three concrete instantiations in (\romannumeral 1) the tabular case; (\romannumeral 2) the case of linear function approximation; and (\romannumeral 3) the case of general function approximation for the outcome model. We then provide a local minimax lower bound to show the instance-dependent optimality of the \textsf{AIPW} estimator using no-regret online learning algorithms.
Authors: Ahmad Sajedi, Samir Khaki, Lucy Z. Liu, Ehsan Amjadian, Yuri A. Lawryshyn, Konstantinos N. Plataniotis
Abstract: Dataset distillation aims to distill the knowledge of a large-scale real dataset into small yet informative synthetic data such that a model trained on it performs as well as a model trained on the full dataset. Despite recent progress, existing dataset distillation methods often struggle with computational efficiency, scalability to complex high-resolution datasets, and generalizability to deep architectures. These approaches typically require retraining when the distillation ratio changes, as knowledge is embedded in raw pixels. In this paper, we propose a novel framework called Data-to-Model Distillation (D2M) to distill the real dataset's knowledge into the learnable parameters of a pre-trained generative model by aligning rich representations extracted from real and generated images. The learned generative model can then produce informative training images for different distillation ratios and deep architectures. Extensive experiments on 15 datasets of varying resolutions show D2M's superior performance, re-distillation efficiency, and cross-architecture generalizability. Our method effectively scales up to high-resolution 128x128 ImageNet-1K. Furthermore, we verify D2M's practical benefits for downstream applications in neural architecture search.
Authors: Vincent Lemaire, Gilles Pag\`es, Christian Yeo
Abstract: We introduce a new class of neural networks designed to be convex functions of their inputs, leveraging the principle that any convex function can be represented as the supremum of the affine functions it dominates. These neural networks, inherently convex with respect to their inputs, are particularly well-suited for approximating the prices of options with convex payoffs. We detail the architecture of this, and establish theoretical convergence bounds that validate its approximation capabilities. We also introduce a \emph{scrambling} phase to improve the training of these networks. Finally, we demonstrate numerically the effectiveness of these networks in estimating prices for three types of options with convex payoffs: Basket, Bermudan, and Swing options.
Authors: Cl\'ement Bonnett, Ariel N. Lee, Franck Wertel, Antoine Tamano, Tanguy Cizain, Pablo Ducru
Abstract: In the last two years, text-to-image diffusion models have become extremely popular. As their quality and usage increase, a major concern has been the need for better output control. In addition to prompt engineering, one effective method to improve the controllability of diffusion models has been to condition them on additional modalities such as image style, depth map, or keypoints. This forms the basis of ControlNets or Adapters. When attempting to apply these methods to control human poses in outputs of text-to-image diffusion models, two main challenges have arisen. The first challenge is generating poses following a wide range of semantic text descriptions, for which previous methods involved searching for a pose within a dataset of (caption, pose) pairs. The second challenge is conditioning image generation on a specified pose while keeping both high aesthetic and high pose fidelity. In this article, we fix these two main issues by introducing a text-to-pose (T2P) generative model alongside a new sampling algorithm, and a new pose adapter that incorporates more pose keypoints for higher pose fidelity. Together, these two new state-of-the-art models enable, for the first time, a generative text-to-pose-to-image framework for higher pose control in diffusion models. We release all models and the code used for the experiments at https://github.com/clement-bonnet/text-to-pose.
Authors: Meryem Altin Karagoz, O. Ufuk Nalbantoglu, Geoffrey C. Fox
Abstract: Deep learning has proven very promising for interpreting MRI in brain tumor diagnosis. However, deep learning models suffer from a scarcity of brain MRI datasets for effective training. Self-supervised learning (SSL) models provide data-efficient and remarkable solutions to limited dataset problems. Therefore, this paper introduces a generative SSL model for brain tumor classification in two stages. The first stage is designed to pre-train a Residual Vision Transformer (ResViT) model for MRI synthesis as a pretext task. The second stage includes fine-tuning a ResViT-based classifier model as a downstream task. Accordingly, we aim to leverage local features via CNN and global features via ViT, employing a hybrid CNN-transformer architecture for ResViT in pretext and downstream tasks. Moreover, synthetic MRI images are utilized to balance the training set. The proposed model performs on public BraTs 2023, Figshare, and Kaggle datasets. Furthermore, we compare the proposed model with various deep learning models, including A-UNet, ResNet-9, pix2pix, pGAN for MRI synthesis, and ConvNeXtTiny, ResNet101, DenseNet12, Residual CNN, ViT for classification. According to the results, the proposed model pretraining on the MRI dataset is superior compared to the pretraining on the ImageNet dataset. Overall, the proposed model attains the highest accuracy, achieving 90.56% on the BraTs dataset with T1 sequence, 98.53% on the Figshare, and 98.47% on the Kaggle brain tumor datasets. As a result, the proposed model demonstrates a robust, effective, and successful approach to handling insufficient dataset challenges in MRI analysis by incorporating SSL, fine-tuning, data augmentation, and combining CNN and ViT.
Authors: Seok-Jin Kim, Min-hwan Oh
Abstract: We study the performance guarantees of exploration-free greedy algorithms for the linear contextual bandit problem. We introduce a novel condition, named the \textit{Local Anti-Concentration} (LAC) condition, which enables a greedy bandit algorithm to achieve provable efficiency. We show that the LAC condition is satisfied by a broad class of distributions, including Gaussian, exponential, uniform, Cauchy, and Student's~$t$ distributions, along with other exponential family distributions and their truncated variants. This significantly expands the class of distributions under which greedy algorithms can perform efficiently. Under our proposed LAC condition, we prove that the cumulative expected regret of the greedy algorithm for the linear contextual bandit is bounded by $O(\operatorname{poly} \log T)$. Our results establish the widest range of distributions known to date that allow a sublinear regret bound for greedy algorithms, further achieving a sharp poly-logarithmic regret.
Authors: Koh Sakano, Kairi Furui, Masahito Ohue
Abstract: Natural products are substances produced by organisms in nature and often possess biological activity and structural diversity. Drug development based on natural products has been common for many years. However, the intricate structures of these compounds present challenges in terms of structure determination and synthesis, particularly compared to the efficiency of high-throughput screening of synthetic compounds. In recent years, deep learning-based methods have been applied to the generation of molecules. In this study, we trained chemical language models on a natural product dataset and generated natural product-like compounds. The results showed that the distribution of the compounds generated was similar to that of natural products. We also evaluated the effectiveness of the generated compounds as drug candidates. Our method can be used to explore the vast chemical space and reduce the time and cost of drug discovery of natural products.
Authors: Thomas Flynn, Patrick Johnstone, Shinjae Yoo
Abstract: In distributed optimization, the communication of model updates can be a performance bottleneck. Consequently, gradient compression has been proposed as a means of increasing optimization throughput. In general, due to information loss, compression introduces a penalty on the number of iterations needed to reach a solution. In this work, we investigate how the iteration penalty depends on the interaction between compression and problem structure, in the context of non-convex stochastic optimization. We focus on linear compression schemes, where compression and decompression can be modeled as multiplication with a random matrix. We consider several distributions of matrices, among them random orthogonal matrices and matrices with random Gaussian entries. We find that in each case, the impact of compression on convergence can be quantified in terms of the norm of the Hessian of the objective, using a norm defined by the compression scheme. The analysis reveals that in certain cases, compression performance is related to low-rank structure or other spectral properties of the problem. In these cases, our bounds predict that the penalty introduced by compression is significantly reduced compared to worst-case bounds that only consider the compression level, ignoring problem data. We verify the theoretical findings on several optimization problems, including fine-tuning an image classification model.
Authors: Eta Yang
Abstract: Sign language translation, especially in gloss-free paradigm, is confronting a dilemma of impracticality and unsustainability due to growing resource-intensive methodologies. Contemporary state-of-the-arts (SOTAs) have significantly hinged on pretrained sophiscated backbones such as Large Language Models (LLMs), embedding sources, or extensive datasets, inducing considerable parametric and computational inefficiency for sustainable use in real-world scenario. Despite their success, following this research direction undermines the overarching mission of this domain to create substantial value to bridge hard-hearing and common populations. Committing to the prevailing trend of LLM and Natural Language Processing (NLP) studies, we pursue a profound essential change in architecture to achieve ground-up improvements without external aid from pretrained models, prior knowledge transfer, or any NLP strategies considered not-from-scratch. Introducing Signformer, a from-scratch Feather-Giant transforming the area towards Edge AI that redefines extremities of performance and efficiency with LLM-competence and edgy-deployable compactness. In this paper, we present nature analysis of sign languages to inform our algorithmic design and deliver a scalable transformer pipeline with convolution and attention novelty. We achieve new 2nd place on leaderboard with a parametric reduction of 467-1807x against the finests as of 2024 and outcompete almost every other methods in a lighter configuration of 0.57 million parameters.
Authors: Asad Aali, Marius Arvinte, Sidharth Kumar, Yamin I. Arefeen, Jonathan I. Tamir
Abstract: We examine the effect of incorporating self-supervised denoising as a pre-processing step for training deep learning (DL) based reconstruction methods on data corrupted by Gaussian noise. K-space data employed for training are typically multi-coil and inherently noisy. Although DL-based reconstruction methods trained on fully sampled data can enable high reconstruction quality, obtaining large, noise-free datasets is impractical. We leverage Generalized Stein's Unbiased Risk Estimate (GSURE) for denoising. We evaluate two DL-based reconstruction methods: Diffusion Probabilistic Models (DPMs) and Model-Based Deep Learning (MoDL). We evaluate the impact of denoising on the performance of these DL-based methods in solving accelerated multi-coil magnetic resonance imaging (MRI) reconstruction. The experiments were carried out on T2-weighted brain and fat-suppressed proton-density knee scans. We observed that self-supervised denoising enhances the quality and efficiency of MRI reconstructions across various scenarios. Specifically, employing denoised images rather than noisy counterparts when training DL networks results in lower normalized root mean squared error (NRMSE), higher structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR) across different SNR levels, including 32dB, 22dB, and 12dB for T2-weighted brain data, and 24dB, 14dB, and 4dB for fat-suppressed knee data. Overall, we showed that denoising is an essential pre-processing technique capable of improving the efficacy of DL-based MRI reconstruction methods under diverse conditions. By refining the quality of input data, denoising can enable the training of more effective DL networks, potentially bypassing the need for noise-free reference MRI scans.
Authors: Wannita Takerngsaksiri, Jirat Pasuksmit, Patanamon Thongtanunam, Chakkrit Tantithamthavorn, Ruixiong Zhang, Fan Jiang, Jing Li, Evan Cook, Kun Chen, Ming Wu
Abstract: Recently, Large Language Models (LLMs)-based multi-agent paradigms for software engineering are introduced to automatically resolve software development tasks (e.g., from a given issue to source code). However, existing work is evaluated based on historical benchmark datasets, does not consider human feedback at each stage of the automated software development process, and has not been deployed in practice. In this paper, we introduce a Human-in-the-loop LLM-based Agents framework (HULA) for software development that allows software engineers to refine and guide LLMs when generating coding plans and source code for a given task. We design, implement, and deploy the HULA framework into Atlassian JIRA for internal uses. Through a multi-stage evaluation of the HULA framework, Atlassian software engineers perceive that HULA can minimize the overall development time and effort, especially in initiating a coding plan and writing code for straightforward tasks. On the other hand, challenges around code quality are raised to be solved in some cases. We draw lessons learned and discuss opportunities for future work, which will pave the way for the advancement of LLM-based agents in software development.
Authors: Samuel Filgueira da Silva, Mehmet Fatih Ozkan, Faissal El Idrissi, Prashanth Ramesh, Marcello Canova
Abstract: Accurate modeling of lithium ion (li-ion) batteries is essential for enhancing the safety, and efficiency of electric vehicles and renewable energy systems. This paper presents a data-inspired approach for improving the fidelity of reduced-order li-ion battery models. The proposed method combines a Genetic Algorithm with Sequentially Thresholded Ridge Regression (GA-STRidge) to identify and compensate for discrepancies between a low-fidelity model (LFM) and data generated either from testing or a high-fidelity model (HFM). The hybrid model, combining physics-based and data-driven methods, is tested across different driving cycles to demonstrate the ability to significantly reduce the voltage prediction error compared to the baseline LFM, while preserving computational efficiency. The model robustness is also evaluated under various operating conditions, showing low prediction errors and high Pearson correlation coefficients for terminal voltage in unseen environments.
Authors: Jake Buzhardt, C. Ricardo Constante-Amores, Michael D. Graham
Abstract: This work explores the relationship between state space methods and Koopman operator-based methods for predicting the time-evolution of nonlinear dynamical systems. We demonstrate that extended dynamic mode decomposition with dictionary learning (EDMD-DL), when combined with a state space projection, is equivalent to a neural network representation of the nonlinear discrete-time flow map on the state space. We highlight how this projection step introduces nonlinearity into the evolution equations, enabling significantly improved EDMD-DL predictions. With this projection, EDMD-DL leads to a nonlinear dynamical system on the state space, which can be represented in either discrete or continuous time. This system has a natural structure for neural networks, where the state is first expanded into a high dimensional feature space followed by a linear mapping which represents the discrete-time map or the vector field as a linear combination of these features. Inspired by these observations, we implement several variations of neural ordinary differential equations (ODEs) and EDMD-DL, developed by combining different aspects of their respective model structures and training procedures. We evaluate these methods using numerical experiments on chaotic dynamics in the Lorenz system and a nine-mode model of turbulent shear flow, showing comparable performance across methods in terms of short-time trajectory prediction, reconstruction of long-time statistics, and prediction of rare events. We also show that these methods provide comparable performance to a non-Markovian approach in terms of prediction of extreme events.
Authors: Wassim El Ahmar, Dhanvin Kolhatkar, Farzan Nowruzi, Robert Laganiere
Abstract: Multiple Object Tracking (MOT) in thermal imaging presents unique challenges due to the lack of visual features and the complexity of motion patterns. This paper introduces an innovative approach to improve MOT in the thermal domain by developing a novel box association method that utilizes both thermal object identity and motion similarity. Our method merges thermal feature sparsity and dynamic object tracking, enabling more accurate and robust MOT performance. Additionally, we present a new dataset comprised of a large-scale collection of thermal and RGB images captured in diverse urban environments, serving as both a benchmark for our method and a new resource for thermal imaging. We conduct extensive experiments to demonstrate the superiority of our approach over existing methods, showing significant improvements in tracking accuracy and robustness under various conditions. Our findings suggest that incorporating thermal identity with motion data enhances MOT performance. The newly collected dataset and source code is available at https://github.com/wassimea/thermalMOT
Authors: Gabriel Chua, Shing Yee Chan, Shaun Khoo
Abstract: Large Language Models are prone to off-topic misuse, where users may prompt these models to perform tasks beyond their intended scope. Current guardrails, which often rely on curated examples or custom classifiers, suffer from high false-positive rates, limited adaptability, and the impracticality of requiring real-world data that is not available in pre-production. In this paper, we introduce a flexible, data-free guardrail development methodology that addresses these challenges. By thoroughly defining the problem space qualitatively and passing this to an LLM to generate diverse prompts, we construct a synthetic dataset to benchmark and train off-topic guardrails that outperform heuristic approaches. Additionally, by framing the task as classifying whether the user prompt is relevant with respect to the system prompt, our guardrails effectively generalize to other misuse categories, including jailbreak and harmful prompts. Lastly, we further contribute to the field by open-sourcing both the synthetic dataset and the off-topic guardrail models, providing valuable resources for developing guardrails in pre-production environments and supporting future research and development in LLM safety.
Authors: Tathagata Sadhukhan, Manit Paul, Raaz Dwivedi
Abstract: Nearest neighbor (NN) algorithms have been extensively used for missing data problems in recommender systems and sequential decision-making systems. Prior theoretical analysis has established favorable guarantees for NN when the underlying data is sufficiently smooth and the missingness probabilities are lower bounded. Here we analyze NN with non-smooth non-linear functions with vast amounts of missingness. In particular, we consider matrix completion settings where the entries of the underlying matrix follow a latent non-linear factor model, with the non-linearity belonging to a \Holder function class that is less smooth than Lipschitz. Our results establish following favorable properties for a suitable two-sided NN: (1) The mean squared error (MSE) of NN adapts to the smoothness of the non-linearity, (2) under certain regularity conditions, the NN error rate matches the rate obtained by an oracle equipped with the knowledge of both the row and column latent factors, and finally (3) NN's MSE is non-trivial for a wide range of settings even when several matrix entries might be missing deterministically. We support our theoretical findings via extensive numerical simulations and a case study with data from a mobile health study, HeartSteps.
Authors: Skyler Seto, Maartje ter Hoeve, He Bai, Natalie Schluter, David Grangier
Abstract: Large language models are trained on massive scrapes of the web, as required by current scaling laws. Most progress is made for English, given its abundance of high-quality pretraining data. For most other languages, however, such high quality pretraining data is unavailable. In this work, we study how to boost pretrained model performance in a data constrained target language by enlisting data from an auxiliary language for which high quality data is available. We study this by quantifying the performance gap between training with data in a data-rich auxiliary language compared with training in the target language, exploring the benefits of translation systems, studying the limitations of model scaling for data constrained languages, and proposing new methods for upsampling data from the auxiliary language. Our results show that stronger auxiliary datasets result in performance gains without modification to the model or training objective for close languages, and, in particular, that performance gains due to the development of more information-rich English pretraining datasets can extend to targeted language settings with limited data.
Authors: Anka Reuel, Amelia Hardy, Chandler Smith, Max Lamparth, Malcolm Hardy, Mykel J. Kochenderfer
Abstract: AI models are increasingly prevalent in high-stakes environments, necessitating thorough assessment of their capabilities and risks. Benchmarks are popular for measuring these attributes and for comparing model performance, tracking progress, and identifying weaknesses in foundation and non-foundation models. They can inform model selection for downstream tasks and influence policy initiatives. However, not all benchmarks are the same: their quality depends on their design and usability. In this paper, we develop an assessment framework considering 46 best practices across an AI benchmark's lifecycle and evaluate 24 AI benchmarks against it. We find that there exist large quality differences and that commonly used benchmarks suffer from significant issues. We further find that most benchmarks do not report statistical significance of their results nor allow for their results to be easily replicated. To support benchmark developers in aligning with best practices, we provide a checklist for minimum quality assurance based on our assessment. We also develop a living repository of benchmark assessments to support benchmark comparability, accessible at betterbench.stanford.edu.
Authors: Zehao Li, Yijie Peng
Abstract: This article addresses the challenge of parameter calibration in stochastic models where the likelihood function is not analytically available. We propose a gradient-based simulated parameter estimation framework, leveraging a multi-time scale algorithm that tackles the issue of ratio bias in both maximum likelihood estimation and posterior density estimation problems. Additionally, we introduce a nested simulation optimization structure, providing theoretical analyses including strong convergence, asymptotic normality, convergence rate, and budget allocation strategies for the proposed algorithm. The framework is further extended to neural network training, offering a novel perspective on stochastic approximation in machine learning. Numerical experiments show that our algorithm can improve the estimation accuracy and save computational costs.
Authors: Haifeng Wen, Nicol\`o Michelusi, Osvaldo Simeone, Hong Xing
Abstract: Over-the-air federated learning (FL), i.e., AirFL, leverages computing primitively over multiple access channels. A long-standing challenge in AirFL is to achieve coherent signal alignment without relying on expensive channel estimation and feedback. This paper proposes NCAirFL, a CSI-free AirFL scheme based on unbiased non-coherent detection at the edge server. By exploiting binary dithering and a long-term memory based error-compensation mechanism, NCAirFL achieves a convergence rate of order $\mathcal{O}(1/\sqrt{T})$ in terms of the average square norm of the gradient for general non-convex and smooth objectives, where $T$ is the number of communication rounds. Experiments demonstrate the competitive performance of NCAirFL compared to vanilla FL with ideal communications and to coherent transmission-based benchmarks.
Authors: Ya\c{s}ar Utku Al\c{c}alar, Merve G\"ulle, Mehmet Ak\c{c}akaya
Abstract: Physics-driven deep learning (PD-DL) approaches have become popular for improved reconstruction of fast magnetic resonance imaging (MRI) scans. Even though PD-DL offers higher acceleration rates compared to existing clinical fast MRI techniques, their use has been limited outside specialized MRI centers. One impediment for their deployment is the difficulties with generalization to pathologies or population groups that are not well-represented in training sets. This has been noted in several studies, and fine-tuning on target populations to improve reconstruction has been suggested. However, current approaches for PD-DL training require access to raw k-space measurements, which is typically only available at specialized MRI centers that have research agreements for such data access. This is especially an issue for rural and underserved areas, where commercial MRI scanners only provide access to a final reconstructed image. To tackle these challenges, we propose Compressibility-inspired Unsupervised Learning via Parallel Imaging Fidelity (CUPID) for high-quality PD-DL training, using only routine clinical reconstructed images exported from an MRI scanner. CUPID evaluates the goodness of the output with a compressibility-based approach, while ensuring that the output stays consistent with the clinical parallel imaging reconstruction through well-designed perturbations. Our results show that CUPID achieves similar quality compared to well-established PD-DL training strategies that require raw k-space data access, while outperforming conventional compressed sensing (CS) and state-of-the-art generative methods. We also demonstrate its effectiveness in a zero-shot training setup for retrospectively and prospectively sub-sampled acquisitions, attesting to its minimal training burden.
Authors: Hung Vinh Tran, Tong Chen, Guanhua Ye, Quoc Viet Hung Nguyen, Kai Zheng, Hongzhi Yin
Abstract: Content-based Recommender Systems (CRSs) play a crucial role in shaping user experiences in e-commerce, online advertising, and personalized recommendations. However, due to the vast amount of categorical features, the embedding tables used in CRS models pose a significant storage bottleneck for real-world deployment, especially on resource-constrained devices. To address this problem, various embedding pruning methods have been proposed, but most existing ones require expensive retraining steps for each target parameter budget, leading to enormous computation costs. In reality, this computation cost is a major hurdle in real-world applications with diverse storage requirements, such as federated learning and streaming settings. In this paper, we propose Shapley Value-guided Embedding Reduction (Shaver) as our response. With Shaver, we view the problem from a cooperative game perspective, and quantify each embedding parameter's contribution with Shapley values to facilitate contribution-based parameter pruning. To address the inherently high computation costs of Shapley values, we propose an efficient and unbiased method to estimate Shapley values of a CRS's embedding parameters. Moreover, in the pruning stage, we put forward a field-aware codebook to mitigate the information loss in the traditional zero-out treatment. Through extensive experiments on three real-world datasets, Shaver has demonstrated competitive performance with lightweight recommendation models across various parameter budgets. The source code is available at https://anonymous.4open.science/r/shaver-E808
Authors: Yifei Zhang, Tianxu Jiang, Bo Pan, Jingyu Wang, Guangji Bai, Liang Zhao
Abstract: Explaining the decision-making processes of Artificial Intelligence (AI) models is crucial for addressing their "black box" nature, particularly in tasks like image classification. Traditional eXplainable AI (XAI) methods typically rely on unimodal explanations, either visual or textual, each with inherent limitations. Visual explanations highlight key regions but often lack rationale, while textual explanations provide context without spatial grounding. Further, both explanation types can be inconsistent or incomplete, limiting their reliability. To address these challenges, we propose a novel Multimodal Explanation-Guided Learning (MEGL) framework that leverages both visual and textual explanations to enhance model interpretability and improve classification performance. Our Saliency-Driven Textual Grounding (SDTG) approach integrates spatial information from visual explanations into textual rationales, providing spatially grounded and contextually rich explanations. Additionally, we introduce Textual Supervision on Visual Explanations to align visual explanations with textual rationales, even in cases where ground truth visual annotations are missing. A Visual Explanation Distribution Consistency loss further reinforces visual coherence by aligning the generated visual explanations with dataset-level patterns, enabling the model to effectively learn from incomplete multimodal supervision. We validate MEGL on two new datasets, Object-ME and Action-ME, for image classification with multimodal explanations. Experimental results demonstrate that MEGL outperforms previous approaches in prediction accuracy and explanation quality across both visual and textual domains. Our code will be made available upon the acceptance of the paper.
Authors: Yijie Zhang, Luzhe Huang, Nir Pillar, Yuzhu Li, Lukasz G. Migas, Raf Van de Plas, Jeffrey M. Spraggins, Aydogan Ozcan
Abstract: Imaging mass spectrometry (IMS) is a powerful tool for untargeted, highly multiplexed molecular mapping of tissue in biomedical research. IMS offers a means of mapping the spatial distributions of molecular species in biological tissue with unparalleled chemical specificity and sensitivity. However, most IMS platforms are not able to achieve microscopy-level spatial resolution and lack cellular morphological contrast, necessitating subsequent histochemical staining, microscopic imaging and advanced image registration steps to enable molecular distributions to be linked to specific tissue features and cell types. Here, we present a virtual histological staining approach that enhances spatial resolution and digitally introduces cellular morphological contrast into mass spectrometry images of label-free human tissue using a diffusion model. Blind testing on human kidney tissue demonstrated that the virtually stained images of label-free samples closely match their histochemically stained counterparts (with Periodic Acid-Schiff staining), showing high concordance in identifying key renal pathology structures despite utilizing IMS data with 10-fold larger pixel size. Additionally, our approach employs an optimized noise sampling technique during the diffusion model's inference process to reduce variance in the generated images, yielding reliable and repeatable virtual staining. We believe this virtual staining method will significantly expand the applicability of IMS in life sciences and open new avenues for mass spectrometry-based biomedical research.
Authors: Hyun Ryu, Eric Kim
Abstract: Efficient inference in large language models (LLMs) has become a critical focus as their scale and complexity grow. Traditional autoregressive decoding, while effective, suffers from computational inefficiencies due to its sequential token generation process. Speculative decoding addresses this bottleneck by introducing a two-stage framework: drafting and verification. A smaller, efficient model generates a preliminary draft, which is then refined by a larger, more sophisticated model. This paper provides a comprehensive survey of speculative decoding methods, categorizing them into draft-centric and model-centric approaches. We discuss key ideas associated with each method, highlighting their potential for scaling LLM inference. This survey aims to guide future research in optimizing speculative decoding and its integration into real-world LLM applications.
Authors: Ofir Cohen, Gil Ari Agmon, Asaf Shabtai, Rami Puzis
Abstract: The popularity of large language models (LLMs) continues to increase, and LLM-based assistants have become ubiquitous, assisting people of diverse backgrounds in many aspects of life. Significant resources have been invested in the safety of LLMs and their alignment with social norms. However, research examining their behavior from the information security awareness (ISA) perspective is lacking. Chatbots and LLM-based assistants may put unwitting users in harm's way by facilitating unsafe behavior. We observe that the ISA inherent in some of today's most popular LLMs varies significantly, with most models requiring user prompts with a clear security context to utilize their security knowledge and provide safe responses to users. Based on this observation, we created a comprehensive set of 30 scenarios to assess the ISA of LLMs. These scenarios benchmark the evaluated models with respect to all focus areas defined in a mobile ISA taxonomy. Among our findings is that ISA is mildly affected by changing the model's temperature, whereas adjusting the system prompt can substantially impact it. This underscores the necessity of setting the right system prompt to mitigate ISA weaknesses. Our findings also highlight the importance of ISA assessment for the development of future LLM-based assistants.
Authors: Ev\v{z}en Wybitul, Evan Ryan Gunter, Mikhail Seleznyov
Abstract: Using vision-language models (VLMs) as reward models in reinforcement learning holds promise for reducing costs and improving safety. So far, VLM reward models have only been used for goal-oriented tasks, where the agent must reach a particular final outcome. We explore VLMs' potential to supervise tasks that cannot be scored by the final state alone. To this end, we introduce ViSTa, a dataset for evaluating Vision-based understanding of Sequential Tasks. ViSTa comprises over 4,000 videos with step-by-step descriptions in virtual home, Minecraft, and real-world environments. Its novel hierarchical structure -- basic single-step tasks composed into more and more complex sequential tasks -- allows a fine-grained understanding of how well VLMs can judge tasks with varying complexity. To illustrate this, we use ViSTa to evaluate state-of-the-art VLMs, including CLIP, ViCLIP, and GPT-4o. We find that, while they are all good at object recognition, they fail to understand sequential tasks, with only GPT-4o achieving non-trivial performance.
Authors: Murray Shanahan, Beth Singler
Abstract: Contemporary conversational AI systems based on large language models (LLMs) can engage users on a wide variety of topics, including philosophy, spirituality, and religion. Suitably prompted, LLMs can be coaxed into discussing such existentially significant matters as their own putative consciousness and the role of artificial intelligence in the fate of the Cosmos. Here we examine two lengthy conversations of this type. We trace likely sources, both ancient and modern, for the extensive repertoire of images, myths, metaphors, and conceptual esoterica that the language model draws on during these conversations, and foreground the contemporary communities and cultural movements that deploy related motifs, especially in their online activity. Finally, we consider the larger societal impacts of such engagements with LLMs.
Authors: Alexander Tolmachev
Abstract: Determining the maximal density $m_1(\mathbb{R}^2)$ of planar sets without unit distances is a fundamental problem in combinatorial geometry. This paper investigates lower bounds for this quantity. We introduce a novel approach to estimating $m_1(\mathbb{R}^2)$ by reformulating the problem as a Maximal Independent Set (MIS) problem on graphs constructed from flat torus, focusing on periodic sets with respect to two non-collinear vectors. Our experimental results supported by theoretical justifications of proposed method demonstrate that for a sufficiently wide range of parameters this approach does not improve the known lower bound $0.22936 \le m_1(\mathbb{R}^2)$. The best discrete sets found are approximations of Croft's construction. In addition, several open source software packages for MIS problem are compared on this task.
Authors: Umamaheswaran Raman Kumar, Abdur Razzaq Fayjie, Jurgen Hannaert, Patrick Vandewalle
Abstract: Large-scale 2D datasets have been instrumental in advancing machine learning; however, progress in 3D vision tasks has been relatively slow. This disparity is largely due to the limited availability of 3D benchmarking datasets. In particular, creating real-world point cloud datasets for indoor scene semantic segmentation presents considerable challenges, including data collection within confined spaces and the costly, often inaccurate process of per-point labeling to generate ground truths. While synthetic datasets address some of these challenges, they often fail to replicate real-world conditions, particularly the occlusions that occur in point clouds collected from real environments. Existing 3D benchmarking datasets typically evaluate deep learning models under the assumption that training and test data are independently and identically distributed (IID), which affects the models' usability for real-world point cloud segmentation. To address these challenges, we introduce the BelHouse3D dataset, a new synthetic point cloud dataset designed for 3D indoor scene semantic segmentation. This dataset is constructed using real-world references from 32 houses in Belgium, ensuring that the synthetic data closely aligns with real-world conditions. Additionally, we include a test set with data occlusion to simulate out-of-distribution (OOD) scenarios, reflecting the occlusions commonly encountered in real-world point clouds. We evaluate popular point-based semantic segmentation methods using our OOD setting and present a benchmark. We believe that BelHouse3D and its OOD setting will advance research in 3D point cloud semantic segmentation for indoor scenes, providing valuable insights for the development of more generalizable models.
Authors: Julian Strohmayer, Rafael Sterzinger, Matthias W\"odlinger, Martin Kampel
Abstract: Cross-domain generalization is an open problem in WiFi-based sensing due to variations in environments, devices, and subjects, causing domain shifts in channel state information. To address this, we propose Domain-Adversarial Test-Time Adaptation (DATTA), a novel framework combining domain-adversarial training (DAT), test-time adaptation (TTA), and weight resetting to facilitate adaptation to unseen target domains and to prevent catastrophic forgetting. DATTA is integrated into a lightweight, flexible architecture optimized for speed. We conduct a comprehensive evaluation of DATTA, including an ablation study on all key components using publicly available data, and verify its suitability for real-time applications such as human activity recognition. When combining a SotA video-based variant of TTA with WiFi-based DAT and comparing it to DATTA, our method achieves an 8.1% higher F1-Score. The PyTorch implementation of DATTA is publicly available at: https://github.com/StrohmayerJ/DATTA.
Authors: Yunli Wang, Zixuan Yang, Zhen Zhang, Zhiqiang Wang, Jian Yang, Shiyang Wen, Peng Jiang, Kun Gai
Abstract: The scaling law is a notable property of neural network models and has significantly propelled the development of large language models. Scaling laws hold great promise in guiding model design and resource allocation. Recent research increasingly shows that scaling laws are not limited to NLP tasks or Transformer architectures; they also apply to domains such as recommendation. However, there is still a lack of literature on scaling law research in online advertisement retrieval systems. This may be because 1) identifying the scaling law for resource cost and online revenue is often expensive in both time and training resources for large-scale industrial applications, and 2) varying settings for different systems prevent the scaling law from being applied across various scenarios. To address these issues, we propose a lightweight paradigm to identify the scaling law of online revenue and machine cost for a certain online advertisement retrieval scenario with a low experimental cost. Specifically, we focus on a sole factor (FLOPs) and propose an offline metric named R/R* that exhibits a high linear correlation with online revenue for retrieval models. We estimate the machine cost offline via a simulation algorithm. Thus, we can transform most online experiments into low-cost offline experiments. We conduct comprehensive experiments to verify the effectiveness of our proposed metric R/R* and to identify the scaling law in the online advertisement retrieval system of Kuaishou. With the scaling law, we demonstrate practical applications for ROI-constrained model designing and multi-scenario resource allocation in Kuaishou advertising system. To the best of our knowledge, this is the first work to study the scaling laws for online advertisement retrieval of real-world systems, showing great potential for scaling law in advertising system optimization.
Authors: Daniel Ramos, Claudia Mamede, Kush Jain, Paulo Canelas, Catarina Gamboa, Claire Le Goues
Abstract: Large Language Models (LLMs) have become integral to various software engineering tasks, including code generation, bug detection, and repair. To evaluate model performance in these domains, numerous bug benchmarks containing real-world bugs from software projects have been developed. However, a growing concern within the software engineering community is that these benchmarks may not reliably reflect true LLM performance due to the risk of data leakage. Despite this concern, limited research has been conducted to quantify the impact of potential leakage. In this paper, we systematically evaluate popular LLMs to assess their susceptibility to data leakage from widely used bug benchmarks. To identify potential leakage, we use multiple metrics, including a study of benchmark membership within commonly used training datasets, as well as analyses of negative log-likelihood and n-gram accuracy. Our findings show that certain models, in particular codegen-multi, exhibit significant evidence of memorization in widely used benchmarks like Defects4J, while newer models trained on larger datasets like LLaMa 3.1 exhibit limited signs of leakage. These results highlight the need for careful benchmark selection and the adoption of robust metrics to adequately assess models capabilities.
Authors: Muqsit Azeem, Debraj Chakraborty, Sudeep Kanav, Jan Kretinsky
Abstract: Partially Observable Markov Decision Processes (POMDPs) are a fundamental framework for decision-making under uncertainty and partial observability. Since in general optimal policies may require infinite memory, they are hard to implement and often render most problems undecidable. Consequently, finite-memory policies are mostly considered instead. However, the algorithms for computing them are typically very complex, and so are the resulting policies. Facing the need for their explainability, we provide a representation of such policies, both (i) in an interpretable formalism and (ii) typically of smaller size, together yielding higher explainability. To that end, we combine models of Mealy machines and decision trees; the latter describing simple, stationary parts of the policies and the former describing how to switch among them. We design a translation for policies of the finite-state-controller (FSC) form from standard literature and show how our method smoothly generalizes to other variants of finite-memory policies. Further, we identify specific properties of recently used "attractor-based" policies, which allow us to construct yet simpler and smaller representations. Finally, we illustrate the higher explainability in a few case studies.
Authors: Lucie Charlotte Magister, Katherine Metcalf, Yizhe Zhang, Maartje ter Hoeve
Abstract: Large Language Models (LLMs) have quickly become an invaluable assistant for a variety of tasks. However, their effectiveness is constrained by their ability to tailor responses to human preferences and behaviors via personalization. Prior work in LLM personalization has largely focused on style transfer or incorporating small factoids about the user, as knowledge injection remains an open challenge. In this paper, we explore injecting knowledge of prior conversations into LLMs to enable future work on less redundant, personalized conversations. We identify two real-world constraints: (1) conversations are sequential in time and must be treated as such during training, and (2) per-user personalization is only viable in parameter-efficient settings. To this aim, we propose PLUM, a pipeline performing data augmentation for up-sampling conversations as question-answer pairs, that are then used to finetune a low-rank adaptation adapter with a weighted cross entropy loss. Even in this first exploration of the problem, we perform competitively with baselines such as RAG, attaining an accuracy of 81.5% across 100 conversations.
Authors: Benedikt Hartl, Yanbo Zhang, Hananel Hazan, Michael Levin
Abstract: Diffusion Models represent a significant advancement in generative modeling, employing a dual-phase process that first degrades domain-specific information via Gaussian noise and restores it through a trainable model. This framework enables pure noise-to-data generation and modular reconstruction of, images or videos. Concurrently, evolutionary algorithms employ optimization methods inspired by biological principles to refine sets of numerical parameters encoding potential solutions to rugged objective functions. Our research reveals a fundamental connection between diffusion models and evolutionary algorithms through their shared underlying generative mechanisms: both methods generate high-quality samples via iterative refinement on random initial distributions. By employing deep learning-based diffusion models as generative models across diverse evolutionary tasks and iteratively refining diffusion models with heuristically acquired databases, we can iteratively sample potentially better-adapted offspring parameters, integrating them into successive generations of the diffusion model. This approach achieves efficient convergence toward high-fitness parameters while maintaining explorative diversity. Diffusion models introduce enhanced memory capabilities into evolutionary algorithms, retaining historical information across generations and leveraging subtle data correlations to generate refined samples. We elevate evolutionary algorithms from procedures with shallow heuristics to frameworks with deep memory. By deploying classifier-free guidance for conditional sampling at the parameter level, we achieve precise control over evolutionary search dynamics to further specific genotypical, phenotypical, or population-wide traits. Our framework marks a major heuristic and algorithmic transition, offering increased flexibility, precision, and control in evolutionary optimization processes.
Authors: Jiacheng Liang, Zian Wang, Lauren Hong, Shouling Ji, Ting Wang
Abstract: To mitigate the misuse of large language models (LLMs), such as disinformation, automated phishing, and academic cheating, there is a pressing need for the capability of identifying LLM-generated texts. Watermarking emerges as one promising solution: it plants statistical signals into LLMs' generative processes and subsequently verifies whether LLMs produce given texts. Various watermarking methods (``watermarkers'') have been proposed; yet, due to the lack of unified evaluation platforms, many critical questions remain under-explored: i) What are the strengths/limitations of various watermarkers, especially their attack robustness? ii) How do various design choices impact their robustness? iii) How to optimally operate watermarkers in adversarial environments? To fill this gap, we systematize existing LLM watermarkers and watermark removal attacks, mapping out their design spaces. We then develop WaterPark, a unified platform that integrates 10 state-of-the-art watermarkers and 12 representative attacks. More importantly, leveraging WaterPark, we conduct a comprehensive assessment of existing watermarkers, unveiling the impact of various design choices on their attack robustness. For instance, a watermarker's resilience to increasingly intensive attacks hinges on its context dependency. We further explore the best practices to operate watermarkers in adversarial environments. For instance, using a generic detector alongside a watermark-specific detector improves the security of vulnerable watermarkers. We believe our study sheds light on current LLM watermarking techniques while WaterPark serves as a valuable testbed to facilitate future research.
Authors: Gaurav Verma, Rachneet Kaur, Nishan Srishankar, Zhen Zeng, Tucker Balch, Manuela Veloso
Abstract: State-of-the-art multimodal web agents, powered by Multimodal Large Language Models (MLLMs), can autonomously execute many web tasks by processing user instructions and interacting with graphical user interfaces (GUIs). Current strategies for building web agents rely on (i) the generalizability of underlying MLLMs and their steerability via prompting, and (ii) large-scale fine-tuning of MLLMs on web-related tasks. However, web agents still struggle to automate tasks on unseen websites and domains, limiting their applicability to enterprise-specific and proprietary platforms. Beyond generalization from large-scale pre-training and fine-tuning, we propose building agents for few-shot adaptability using human demonstrations. We introduce the AdaptAgent framework that enables both proprietary and open-weights multimodal web agents to adapt to new websites and domains using few human demonstrations (up to 2). Our experiments on two popular benchmarks -- Mind2Web & VisualWebArena -- show that using in-context demonstrations (for proprietary models) or meta-adaptation demonstrations (for meta-learned open-weights models) boosts task success rate by 3.36% to 7.21% over non-adapted state-of-the-art models, corresponding to a relative increase of 21.03% to 65.75%. Furthermore, our additional analyses (a) show the effectiveness of multimodal demonstrations over text-only ones, (b) shed light on the influence of different data selection strategies during meta-learning on the generalization of the agent, and (c) demonstrate the effect of number of few-shot examples on the web agent's success rate. Overall, our results unlock a complementary axis for developing widely applicable multimodal web agents beyond large-scale pre-training and fine-tuning, emphasizing few-shot adaptability.
Authors: Sarbartha Banerjee, Prateek Sahu, Mulong Luo, Anjo Vahldiek-Oberwagner, Neeraja J. Yadwadkar, Mohit Tiwari
Abstract: Large language models (LLMs) used across enterprises often use proprietary models and operate on sensitive inputs and data. The wide range of attack vectors identified in prior research - targeting various software and hardware components used in training and inference - makes it extremely challenging to enforce confidentiality and integrity policies. As we advance towards constructing compound AI inference pipelines that integrate multiple large language models (LLMs), the attack surfaces expand significantly. Attackers now focus on the AI algorithms as well as the software and hardware components associated with these systems. While current research often examines these elements in isolation, we find that combining cross-layer attack observations can enable powerful end-to-end attacks with minimal assumptions about the threat model. Given, the sheer number of existing attacks at each layer, we need a holistic and systemized understanding of different attack vectors at each layer. This SoK discusses different software and hardware attacks applicable to compound AI systems and demonstrates how combining multiple attack mechanisms can reduce the threat model assumptions required for an isolated attack. Next, we systematize the ML attacks in lines with the Mitre Att&ck framework to better position each attack based on the threat model. Finally, we outline the existing countermeasures for both software and hardware layers and discuss the necessity of a comprehensive defense strategy to enable the secure and high-performance deployment of compound AI systems.
Authors: Yunbum Kook, Santosh S. Vempala
Abstract: We study the complexity of sampling, rounding, and integrating arbitrary logconcave functions. Our new approach provides the first complexity improvements in nearly two decades for general logconcave functions for all three problems, and matches the best-known complexities for the special case of uniform distributions on convex bodies. For the sampling problem, our output guarantees are significantly stronger than previously known, and lead to a streamlined analysis of statistical estimation based on dependent random samples.
Authors: Guillaume Principato, Yvenn Amara-Ouali, Yannig Goude, Bachir Hamrouche, Jean-Michel Poggi, Gilles Stoltz
Abstract: Reconciliation has become an essential tool in multivariate point forecasting for hierarchical time series. However, there is still a lack of understanding of the theoretical properties of probabilistic Forecast Reconciliation techniques. Meanwhile, Conformal Prediction is a general framework with growing appeal that provides prediction sets with probabilistic guarantees in finite sample. In this paper, we propose a first step towards combining Conformal Prediction and Forecast Reconciliation by analyzing how including a reconciliation step in the Split Conformal Prediction (SCP) procedure enhances the resulting prediction sets. In particular, we show that the validity granted by SCP remains while improving the efficiency of the prediction sets. We also advocate a variation of the theoretical procedure for practical use. Finally, we illustrate these results with simulations.
Authors: John D. Hastings, Sherri Weitl-Harms, Joseph Doty, Zachary L. Myers, Warren Thompson
Abstract: This research explores the application of large language models (LLMs) to generate synthetic datasets for Product Desirability Toolkit (PDT) testing, a key component in evaluating user sentiment and product experience. Utilizing gpt-4o-mini, a cost-effective alternative to larger commercial LLMs, three methods, Word+Review, Review+Word, and Supply-Word, were each used to synthesize 1000 product reviews. The generated datasets were assessed for sentiment alignment, textual diversity, and data generation cost. Results demonstrated high sentiment alignment across all methods, with Pearson correlations ranging from 0.93 to 0.97. Supply-Word exhibited the highest diversity and coverage of PDT terms, although with increased generation costs. Despite minor biases toward positive sentiments, in situations with limited test data, LLM-generated synthetic data offers significant advantages, including scalability, cost savings, and flexibility in dataset production.
Authors: Yuan Deng, Amin Karbasi, Vahab Mirrokni, Renato Paes Leme, Grigoris Velegkas, Song Zuo
Abstract: We study procurement auctions, where an auctioneer seeks to acquire services from strategic sellers with private costs. The quality of services is measured by a submodular function known to the auctioneer. Our goal is to design computationally efficient procurement auctions that (approximately) maximize the difference between the quality of the acquired services and the total cost of the sellers, while ensuring incentive compatibility (IC), individual rationality (IR) for sellers, and non-negative surplus (NAS) for the auctioneer. Our contributions are twofold: (i) we provide an improved analysis of existing algorithms for non-positive submodular function maximization, and (ii) we design efficient frameworks that transform submodular optimization algorithms into mechanisms that are IC, IR, NAS, and approximation-preserving. These frameworks apply to both the offline setting, where all sellers' bids and services are available simultaneously, and the online setting, where sellers arrive in an adversarial order, requiring the auctioneer to make irrevocable decisions. We also explore whether state-of-the-art submodular optimization algorithms can be converted into descending auctions in adversarial settings, where the schedule of descending prices is determined by an adversary. We show that a submodular optimization algorithm satisfying bi-criteria $(1/2, 1)$-approximation in welfare can be effectively adapted to a descending auction. Additionally, we establish a connection between descending auctions and online submodular optimization. Finally, we demonstrate the practical applications of our frameworks by instantiating them with state-of-the-art submodular optimization algorithms and empirically comparing their welfare performance on publicly available datasets with thousands of sellers.
Authors: Alessandro Tesi, Gopal Ramesh Dahale, Sergei Gleyzer, Kyoungchul Kong, Tom Magorsch, Konstantin T. Matchev, Katia Matcheva
Abstract: We present a novel hybrid quantum-classical vision transformer architecture incorporating quantum orthogonal neural networks (QONNs) to enhance performance and computational efficiency in high-energy physics applications. Building on advancements in quantum vision transformers, our approach addresses limitations of prior models by leveraging the inherent advantages of QONNs, including stability and efficient parameterization in high-dimensional spaces. We evaluate the proposed architecture using multi-detector jet images from CMS Open Data, focusing on the task of distinguishing quark-initiated from gluon-initiated jets. The results indicate that embedding quantum orthogonal transformations within the attention mechanism can provide robust performance while offering promising scalability for machine learning challenges associated with the upcoming High Luminosity Large Hadron Collider. This work highlights the potential of quantum-enhanced models to address the computational demands of next-generation particle physics experiments.
Authors: Bahri Batuhan Bilecen, Ahmet Berke Gokmen, Furkan Guzelant, Aysegul Dundar
Abstract: 3D head stylization transforms realistic facial features into artistic representations, enhancing user engagement across gaming and virtual reality applications. While 3D-aware generators have made significant advancements, many 3D stylization methods primarily provide near-frontal views and struggle to preserve the unique identities of original subjects, often resulting in outputs that lack diversity and individuality. This paper addresses these challenges by leveraging the PanoHead model, synthesizing images from a comprehensive 360-degree perspective. We propose a novel framework that employs negative log-likelihood distillation (LD) to enhance identity preservation and improve stylization quality. By integrating multi-view grid score and mirror gradients within the 3D GAN architecture and introducing a score rank weighing technique, our approach achieves substantial qualitative and quantitative improvements. Our findings not only advance the state of 3D head stylization but also provide valuable insights into effective distillation processes between diffusion models and GANs, focusing on the critical issue of identity preservation. Please visit the https://three-bee.github.io/head_stylization for more visuals.
Authors: Shoaib Meraj Sami, Md Mahedi Hasan, Jeremy Dawson, Nasser Nasrabadi
Abstract: Although recent diffusion-based single-step super-resolution methods achieve better performance as compared to SinSR, they are computationally complex. To improve the performance of SinSR, we investigate preserving the high-frequency detail features during super-resolution (SR) because the downgraded images lack detailed information. For this purpose, we introduce a high-frequency perceptual loss by utilizing an invertible neural network (INN) pretrained on the ImageNet dataset. Different feature maps of pretrained INN produce different high-frequency aspects of an image. During the training phase, we impose to preserve the high-frequency features of super-resolved and ground truth (GT) images that improve the SR image quality during inference. Furthermore, we also utilize the Jenson-Shannon divergence between GT and SR images in the pretrained DINO-v2 embedding space to match their distribution. By introducing the $\textbf{h}igh$- $\textbf{f}requency$ preserving loss and distribution matching constraint in the single-step $\textbf{diff}usion-based$ SR ($\textbf{HF-Diff}$), we achieve a state-of-the-art CLIPIQA score in the benchmark RealSR, RealSet65, DIV2K-Val, and ImageNet datasets. Furthermore, the experimental results in several datasets demonstrate that our high-frequency perceptual loss yields better SR image quality than LPIPS and VGG-based perceptual losses. Our code will be released at https://github.com/shoaib-sami/HF-Diff.
Authors: Moyang Guo, Yuepeng Hu, Zhengyuan Jiang, Zeyu Li, Amir Sadovnik, Arka Daw, Neil Gong
Abstract: While text-to-image models offer numerous benefits, they also pose significant societal risks. Detecting AI-generated images is crucial for mitigating these risks. Detection methods can be broadly categorized into passive and watermark-based approaches: passive detectors rely on artifacts present in AI-generated images, whereas watermark-based detectors proactively embed watermarks into such images. A key question is which type of detector performs better in terms of effectiveness, robustness, and efficiency. However, the current literature lacks a comprehensive understanding of this issue. In this work, we aim to bridge that gap by developing ImageDetectBench, the first comprehensive benchmark to compare the effectiveness, robustness, and efficiency of passive and watermark-based detectors. Our benchmark includes four datasets, each containing a mix of AI-generated and non-AI-generated images. We evaluate five passive detectors and four watermark-based detectors against eight types of common perturbations and three types of adversarial perturbations. Our benchmark results reveal several interesting findings. For instance, watermark-based detectors consistently outperform passive detectors, both in the presence and absence of perturbations. Based on these insights, we provide recommendations for detecting AI-generated images, e.g., when both types of detectors are applicable, watermark-based detectors should be the preferred choice.
Authors: Niharika Shimona D'Souza, Mary Beth Nebel, Nicholas Wymbs, Stewart Mostofsky, Archana Venkataraman
Abstract: We propose a unified optimization framework that combines neural networks with dictionary learning to model complex interactions between resting state functional MRI and behavioral data. The dictionary learning objective decomposes patient correlation matrices into a collection of shared basis networks and subject-specific loadings. These subject-specific features are simultaneously input into a neural network that predicts multidimensional clinical information. Our novel optimization framework combines the gradient information from the neural network with that of a conventional matrix factorization objective. This procedure collectively estimates the basis networks, subject loadings, and neural network weights most informative of clinical severity. We evaluate our combined model on a multi-score prediction task using 52 patients diagnosed with Autism Spectrum Disorder (ASD). Our integrated framework outperforms state-of-the-art methods in a ten-fold cross validated setting to predict three different measures of clinical severity.
Authors: Lijun Zhang, Yibo Wang, Guanghui Wang, Jinfeng Yi, Tianbao Yang
Abstract: Recently, several universal methods have been proposed for online convex optimization, and attain minimax rates for multiple types of convex functions simultaneously. However, they need to design and optimize one surrogate loss for each type of functions, making it difficult to exploit the structure of the problem and utilize existing algorithms. In this paper, we propose a simple strategy for universal online convex optimization, which avoids these limitations. The key idea is to construct a set of experts to process the original online functions, and deploy a meta-algorithm over the linearized losses to aggregate predictions from experts. Specifically, the meta-algorithm is required to yield a second-order bound with excess losses, so that it can leverage strong convexity and exponential concavity to control the meta-regret. In this way, our strategy inherits the theoretical guarantee of any expert designed for strongly convex functions and exponentially concave functions, up to a double logarithmic factor. As a result, we can plug in off-the-shelf online solvers as black-box experts to deliver problem-dependent regret bounds. For general convex functions, it maintains the minimax optimality and also achieves a small-loss bound. Furthermore, we extend our universal strategy to online composite optimization, where the loss function comprises a time-varying function and a fixed regularizer. To deal with the composite loss functions, we employ a meta-algorithm based on the optimistic online learning framework, which not only possesses a second-order bound, but also can utilize estimations for upcoming loss functions. With appropriate configurations, we demonstrate that the additional regularizer does not contribute to the meta-regret, thus maintaining the universality in the composite setting.
Authors: Haibin Zhou, Tong Wei, Zichuan Lin, junyou li, Junliang Xing, Yuanchun Shi, Li Shen, Chao Yu, Deheng Ye
Abstract: We study the adaption of Soft Actor-Critic (SAC), which is considered as a state-of-the-art reinforcement learning (RL) algorithm, from continuous action space to discrete action space. We revisit vanilla discrete SAC and provide an in-depth understanding of its Q value underestimation and performance instability issues when applied to discrete settings. We thereby propose Stable Discrete SAC (SDSAC), an algorithm that leverages entropy-penalty and double average Q-learning with Q-clip to address these issues. Extensive experiments on typical benchmarks with discrete action space, including Atari games and a large-scale MOBA game, show the efficacy of our proposed method. Our code is at: https://github.com/coldsummerday/SD-SAC.git.
Authors: Emmanuel Abbe, Samy Bengio, Aryo Lotfi, Kevin Rizk
Abstract: This paper considers the learning of logical (Boolean) functions with a focus on the generalization on the unseen (GOTU) setting, a strong case of out-of-distribution generalization. This is motivated by the fact that the rich combinatorial nature of data in certain reasoning tasks (e.g., arithmetic/logic) makes representative data sampling challenging, and learning successfully under GOTU gives a first vignette of an 'extrapolating' or 'reasoning' learner. We study how different network architectures trained by (S)GD perform under GOTU and provide both theoretical and experimental evidence that for sparse functions and a class of network models including instances of Transformers, random features models, and linear networks, a min-degree-interpolator is learned on the unseen. More specifically, this means an interpolator of the training data that has minimal Fourier mass on the higher degree basis elements. These findings lead to two implications: (1) we provide an explanation to the length generalization problem for Boolean functions (e.g., Anil et al. 2022); (2) we introduce a curriculum learning algorithm called Degree-Curriculum that learns monomials more efficiently by incrementing supports. Finally, we discuss extensions to other models or non-sparse regimes where the min-degree bias may still occur or fade, as well as how it can be potentially corrected when undesirable.
Authors: Lijun Zhang, Haomin Bai, Peng Zhao, Tianbao Yang, Zhi-Hua Zhou
Abstract: This paper investigates group distributionally robust optimization (GDRO) with the goal of learning a model that performs well over $m$ different distributions. First, we formulate GDRO as a stochastic convex-concave saddle-point problem, which is then solved by stochastic mirror descent (SMD) with $m$ samples in each iteration, and attain a nearly optimal sample complexity. To reduce the number of samples required in each round from $m$ to 1, we cast GDRO as a two-player game, where one player conducts SMD and the other executes an online algorithm for non-oblivious multi-armed bandits, maintaining the same sample complexity. Next, we extend GDRO to address scenarios involving imbalanced data and heterogeneous distributions. In the first scenario, we introduce a weighted variant of GDRO, enabling distribution-dependent convergence rates that rely on the number of samples from each distribution. We design two strategies to meet the sample budget: one integrates non-uniform sampling into SMD, and the other employs the stochastic mirror-prox algorithm with mini-batches, both of which deliver faster rates for distributions with more samples. In the second scenario, we propose to optimize the average top-$k$ risk instead of the maximum risk, thereby mitigating the impact of outlier distributions. Similar to the case of vanilla GDRO, we develop two stochastic approaches: one uses $m$ samples per iteration via SMD, and the other consumes $k$ samples per iteration through an online algorithm for non-oblivious combinatorial semi-bandits.
Authors: Khashayar Khosravi, Renato Paes Leme, Chara Podimata, Apostolis Tsorvantzis
Abstract: We propose a model for learning with bandit feedback while accounting for deterministically evolving and unobservable states that we call Bandits with Deterministically Evolving States ($B$-$DES$). The workhorse applications of our model are learning for recommendation systems and learning for online ads. In both cases, the reward that the algorithm obtains at each round is a function of the short-term reward of the action chosen and how "healthy" the system is (i.e., as measured by its state). For example, in recommendation systems, the reward that the platform obtains from a user's engagement with a particular type of content depends not only on the inherent features of the specific content, but also on how the user's preferences have evolved as a result of interacting with other types of content on the platform. Our general model accounts for the different rate $\lambda \in [0,1]$ at which the state evolves (e.g., how fast a user's preferences shift as a result of previous content consumption) and encompasses standard multi-armed bandits as a special case. The goal of the algorithm is to minimize a notion of regret against the best-fixed sequence of arms pulled, which is significantly harder to attain compared to standard benchmark of the best-fixed action in hindsight. We present online learning algorithms for any possible value of the evolution rate $\lambda$ and we show the robustness of our results to various model misspecifications.
Authors: Eric Hsiung, Joydeep Biswas, Swarat Chaudhuri
Abstract: Active automata learning from membership and equivalence queries is a foundational problem with numerous applications. We propose a novel variant of the active automata learning problem: actively learn finite automata using preference queries -- i.e., queries about the relative position of two sequences in a total order -- instead of membership queries. Our solution is REMAP, a novel algorithm which leverages a symbolic observation table along with unification and constraint solving to navigate a space of symbolic hypotheses (each representing a set of automata), and uses satisfiability-solving to construct a concrete automaton from a symbolic hypothesis. REMAP is guaranteed to correctly infer the minimal automaton with polynomial query complexity under exact equivalence queries, and achieves PAC-identification ($\varepsilon$-approximate, with high probability) of the minimal automaton using sampling-based equivalence queries. Our empirical evaluations of REMAP on the task of learning reward machines for two reinforcement learning domains indicate REMAP scales to large automata and is effective at learning correct automata from consistent teachers, under both exact and sampling-based equivalence queries.
Authors: Nivasini Ananthakrishnan, Stephen Bates, Michael I. Jordan, Nika Haghtalab
Abstract: Motivated by the emergence of decentralized machine learning (ML) ecosystems, we study the delegation of data collection. Taking the field of contract theory as our starting point, we design optimal and near-optimal contracts that deal with two fundamental information asymmetries that arise in decentralized ML: uncertainty in the assessment of model quality and uncertainty regarding the optimal performance of any model. We show that a principal can cope with such asymmetry via simple linear contracts that achieve 1-1/e fraction of the optimal utility. To address the lack of a priori knowledge regarding the optimal performance, we give a convex program that can adaptively and efficiently compute the optimal contract. We also study linear contracts and derive the optimal utility in the more complex setting of multiple interactions.
Authors: Deyu Zou, Shikun Liu, Siqi Miao, Victor Fung, Shiyu Chang, Pan Li
Abstract: Geometric deep learning (GDL) has gained significant attention in scientific fields, for its proficiency in modeling data with intricate geometric structures. However, very few works have delved into its capability of tackling the distribution shift problem, a prevalent challenge in many applications. To bridge this gap, we propose GeSS, a comprehensive benchmark designed for evaluating the performance of GDL models in scientific scenarios with distribution shifts. Our evaluation datasets cover diverse scientific domains from particle physics, materials science to biochemistry, and encapsulate a broad spectrum of distribution shifts including conditional, covariate, and concept shifts. Furthermore, we study three levels of information access from the out-of-distribution (OOD) test data, including no OOD information, only unlabeled OOD data, and OOD data with a few labels. Overall, our benchmark results in 30 different experiment settings, and evaluates 3 GDL backbones and 11 learning algorithms in each setting. A thorough analysis of the evaluation results is provided, poised to illuminate insights for GDL researchers and domain practitioners who are to use GDL in their applications.
Authors: Haoyu Geng, Hang Ruan, Runzhong Wang, Yang Li, Yang Wang, Lei Chen, Junchi Yan
Abstract: Predictive combinatorial optimization, where the parameters of combinatorial optimization (CO) are unknown at the decision-making time, is the precise modeling of many real-world applications, including energy cost-aware scheduling and budget allocation on advertising. Tackling such a problem usually involves a prediction model and a CO solver. These two modules are integrated into the predictive CO pipeline following two design principles: "Predict-then-Optimize (PtO)", which learns predictions by supervised training and subsequently solves CO using predicted coefficients, while the other, named "Predict-and-Optimize (PnO)", directly optimizes towards the ultimate decision quality and claims to yield better decisions than traditional PtO approaches. However, there lacks a systematic benchmark of both approaches, including the specific design choices at the module level, as well as an evaluation dataset that covers representative real-world scenarios. To this end, we develop a modular framework to benchmark 11 existing PtO/PnO methods on 8 problems, including a new industrial dataset for combinatorial advertising that will be released. Our study shows that PnO approaches are better than PtO on 7 out of 8 benchmarks, but there is no silver bullet found for the specific design choices of PnO. A comprehensive categorization of current approaches and integration of typical scenarios are provided under a unified benchmark. Therefore, this paper could serve as a comprehensive benchmark for future PnO approach development and also offer fast prototyping for application-focused development. The code is available at https://github.com/Thinklab-SJTU/PredictiveCO-Benchmark.
URLs: https://github.com/Thinklab-SJTU/PredictiveCO-Benchmark.
Authors: Chollette Olisah, Lyndon Smith, Melvyn Smith, Lawrence Morolake, Osi Ojukwu
Abstract: Crop yield prediction has been modeled on the assumption that there is no interaction between weather and soil variables. However, this paper argues that an interaction exists, and it can be finely modelled using the Kendall Correlation coefficient. Given the nonlinearity of the interaction between weather and soil variables, a deep neural network regressor (DNNR) is carefully designed with consideration to the depth, number of neurons of the hidden layers, and the hyperparameters with their optimizations. Additionally, a new metric, the average of absolute root squared error (ARSE) is proposed to combine the strengths of root mean square error (RMSE) and mean absolute error (MAE). With the ARSE metric, the proposed DNNR(s), optimised random forest regressor (RFR) and the extreme gradient boosting regressor (XGBR) achieved impressively small yield errors, 0.0172 t/ha, and 0.0243 t/ha, 0.0001 t/ha, and 0.001 t/ha, respectively. However, the DNNR(s), with changes to the explanatory variables to ensure generalizability to unforeseen data, DNNR(s) performed best. Further analysis reveals that a strong interaction does exist between weather and soil variables. Precisely, yield is observed to increase when precipitation is reduced and silt increased, and vice-versa. However, the degree of decrease or increase is not quantified in this paper. Contrary to existing yield models targeted towards agricultural policies and global food security, the goal of the proposed corn yield model is to empower the smallholder farmer to farm smartly and intelligently, thus the prediction model is integrated into a mobile application that includes education, and a farmer-to-market access module.
Authors: Andr\'e Kelm, Niels Hannemann, Bruno Heberle, Lucas Schmidt, Tim Rolff, Christian Wilms, Ehsan Yaghoubi, Simone Frintrop
Abstract: This study introduces a novel expert generation method that dynamically reduces task and computational complexity without compromising predictive performance. It is based on a new hierarchical classification network topology that combines sequential processing of generic low-level features with parallelism and nesting of high-level features. This structure allows for the innovative extraction technique: the ability to select only high-level features of task-relevant categories. In certain cases, it is possible to skip almost all unneeded high-level features, which can significantly reduce the inference cost and is highly beneficial in resource-constrained conditions. We believe this method paves the way for future network designs that are lightweight and adaptable, making them suitable for a wide range of applications, from compact edge devices to large-scale clouds. In terms of dynamic inference our methodology can achieve an exclusion of up to 88.7\,\% of parameters and 73.4\,\% fewer giga-multiply accumulate (GMAC) operations, analysis against comparative baselines showing an average reduction of 47.6\,\% in parameters and 5.8\,\% in GMACs across the cases we evaluated.
Authors: Gijs Bellaard, Sei Sakata, Bart M. N. Smets, Remco Duits
Abstract: PDE-based Group Convolutional Neural Networks (PDE-G-CNNs) use solvers of evolution PDEs as substitutes for the conventional components in G-CNNs. PDE-G-CNNs can offer several benefits simultaneously: fewer parameters, inherent equivariance, better accuracy, and data efficiency. In this article we focus on Euclidean equivariant PDE-G-CNNs where the feature maps are two-dimensional throughout. We call this variant of the framework a PDE-CNN. From a machine learning perspective, we list several practically desirable axioms and derive from these which PDEs should be used in a PDE-CNN, this being our main contribution. Our approach to geometric learning via PDEs is inspired by the axioms of scale-space theory, which we generalize by introducing semifield-valued signals. Our theory reveals new PDEs that can be used in PDE-CNNs and we experimentally examine what impact these have on the accuracy of PDE-CNNs. We also confirm for small networks that PDE-CNNs offer fewer parameters, increased accuracy, and better data efficiency when compared to CNNs.
Authors: Kafeng Wang, Jianfei Chen, He Li, Zhenpeng Mi, Jun Zhu
Abstract: Diffusion models have been extensively used in data generation tasks and are recognized as one of the best generative models. However, their time-consuming deployment, long inference time, and requirements on large memory limit their application on mobile devices. In this paper, we propose a method based on the improved Straight-Through Estimator to improve the deployment efficiency of diffusion models. Specifically, we add sparse masks to the Convolution and Linear layers in a pre-trained diffusion model, then use design progressive sparsity for model training in the fine-tuning stage, and switch the inference mask on and off, which supports a flexible choice of sparsity during inference according to the FID and MACs requirements. Experiments on four datasets conducted on a state-of-the-art Transformer-based diffusion model demonstrate that our method reduces MACs by $50\%$ while increasing FID by only 1.5 on average. Under other MACs conditions, the FID is also lower than 1$\sim$137 compared to other methods.
Authors: Skander Moalla, Andrea Miele, Daniil Pyatko, Razvan Pascanu, Caglar Gulcehre
Abstract: Reinforcement learning (RL) is inherently rife with non-stationarity since the states and rewards the agent observes during training depend on its changing policy. Therefore, networks in deep RL must be capable of adapting to new observations and fitting new targets. However, previous works have observed that networks trained under non-stationarity exhibit an inability to continue learning, termed loss of plasticity, and eventually a collapse in performance. For off-policy deep value-based RL methods, this phenomenon has been correlated with a decrease in representation rank and the ability to fit random targets, termed capacity loss. Although this correlation has generally been attributed to neural network learning under non-stationarity, the connection to representation dynamics has not been carefully studied in on-policy policy optimization methods. In this work, we empirically study representation dynamics in Proximal Policy Optimization (PPO) on the Atari and MuJoCo environments, revealing that PPO agents are also affected by feature rank deterioration and capacity loss. We show that this is aggravated by stronger non-stationarity, ultimately driving the actor's performance to collapse, regardless of the performance of the critic. We ask why the trust region, specific to methods like PPO, cannot alleviate or prevent the collapse and find a connection between representation collapse and the degradation of the trust region, one exacerbating the other. Finally, we present Proximal Feature Optimization (PFO), a novel auxiliary loss that, along with other interventions, shows that regularizing the representation dynamics mitigates the performance collapse of PPO agents.
Authors: Marcin Podhajski, Jan Dubi\'nski, Franziska Boenisch, Adam Dziedzic, Agnieszka Pregowska, Tomasz P. Michalak
Abstract: Graph Neural Networks (GNNs) are recognized as potent tools for processing real-world data organized in graph structures. Especially inductive GNNs, which allow for the processing of graph-structured data without relying on predefined graph structures, are becoming increasingly important in a wide range of applications. As such these networks become attractive targets for model-stealing attacks where an adversary seeks to replicate the functionality of the targeted network. Significant efforts have been devoted to developing model-stealing attacks that extract models trained on images and texts. However, little attention has been given to stealing GNNs trained on graph data. This paper identifies a new method of performing unsupervised model-stealing attacks against inductive GNNs, utilizing graph contrastive learning and spectral graph augmentations to efficiently extract information from the targeted model. The new type of attack is thoroughly evaluated on six datasets and the results show that our approach outperforms the current state-of-the-art by Shen et al. (2021). In particular, our attack surpasses the baseline across all benchmarks, attaining superior fidelity and downstream accuracy of the stolen model while necessitating fewer queries directed toward the target model.
Authors: Xiachong Lin, Arian Prabowo, Imran Razzak, Hao Xue, Matthew Amos, Sam Behrens, Flora D. Salim
Abstract: The increasing demand for sustainable energy solutions has driven the integration of digitalized buildings into the power grid, leveraging Internet-of-Things (IoT) technologies to enhance energy efficiency and operational performance. Despite their potential, effectively utilizing IoT point data within deep-learning frameworks presents significant challenges, primarily due to its inherent heterogeneity. This study investigates the diverse dimensions of IoT data heterogeneity in both intra-building and inter-building contexts, examining their implications for predictive modeling. A benchmarking analysis of state-of-the-art time series models highlights their performance on this complex dataset. The results emphasize the critical need for multi-modal data integration, domain-informed modeling, and automated data engineering pipelines. Additionally, the study advocates for collaborative efforts to establish high-quality public datasets, which are essential for advancing intelligent and sustainable energy management systems in digitalized buildings.
Authors: Sharmila Duppala, Juan Luque, John P. Dickerson, Seyed A. Esmaeili
Abstract: We study the canonical fair clustering problem where each cluster is constrained to have close to population-level representation of each group. Despite significant attention, the salient issue of having incomplete knowledge about the group membership of each point has been superficially addressed. In this paper, we consider a setting where the assigned group memberships are noisy. We introduce a simple noise model that requires a small number of parameters to be given by the decision maker. We then present an algorithm for fair clustering with provable \emph{robustness} guarantees. Our framework enables the decision maker to trade off between the robustness and the clustering quality. Unlike previous work, our algorithms are backed by worst-case theoretical guarantees. Finally, we empirically verify the performance of our algorithm on real world datasets and show its superior performance over existing baselines.
Authors: AmirPouya Hemmasian, Amir Barati Farimani
Abstract: There has recently been increasing attention towards developing foundational neural Partial Differential Equation (PDE) solvers and neural operators through large-scale pretraining. However, unlike vision and language models that make use of abundant and inexpensive (unlabeled) data for pretraining, these neural solvers usually rely on simulated PDE data, which can be costly to obtain, especially for high-dimensional PDEs. In this work, we aim to Pretrain neural PDE solvers on Lower Dimensional PDEs (PreLowD) where data collection is the least expensive. We evaluated the effectiveness of this pretraining strategy in similar PDEs in higher dimensions. We use the Factorized Fourier Neural Operator (FFNO) due to having the necessary flexibility to be applied to PDE data of arbitrary spatial dimensions and reuse trained parameters in lower dimensions. In addition, our work sheds light on the effect of the fine-tuning configuration to make the most of this pretraining strategy. Code is available at https://github.com/BaratiLab/PreLowD.
Authors: Juyoung Yun
Abstract: The rapid advancements in deep learning necessitate better training methods for deep neural networks (DNNs). As models grow in complexity, vanishing and exploding gradients impede performance, particularly in skip-connected architectures like Deep Residual Networks. We propose Z-Score Normalization for Gradient Descent (ZNorm), an innovative technique that adjusts only the gradients without modifying the network architecture to accelerate training and improve model performance. ZNorm normalizes the overall gradients, providing consistent gradient scaling across layers, effectively reducing the risks of vanishing and exploding gradients and achieving superior performance. Extensive experiments on CIFAR-10 and medical datasets confirm that ZNorm consistently outperforms existing methods under the same experimental settings. In medical imaging applications, ZNorm significantly enhances tumor prediction and segmentation accuracy, underscoring its practical utility. These findings highlight ZNorm's potential as a robust and versatile tool for enhancing the training and effectiveness of deep neural networks, especially in skip-connected architectures, across various applications.
Authors: Yang Cao
Abstract: In this paper, we propose Singular Values and Orthonormal Regularized Singular Vectors Adaptation, or SORSA, a novel PEFT method. Each SORSA adapter consists of two main parts: trainable principal singular weights $W_p = U_p \text{diag}(S_p) V^\top_p$, and frozen residual weights $W_r = U_r \text{diag}(S_r) V^\top_r$. These parts are initialized by performing singular value decomposition (SVD) on pre-trained weights. Moreover, we implement and analyze an orthonormal regularizer, which we prove could decrease the condition number of $W_p$ and make the optimization more efficient. SORSA adapters could be merged during inference, thus eliminating any inference latency. We also introduce a method to analyze the variation of the parameters by performing SVD and discuss and analyze SORSA's superiority in minimizing the alteration in the SVD aspect. After all, SORSA shows a faster convergence than LoRA and PiSSA in our experiments. On the GSM-8K benchmark, Llama 2 7B adapted using SORSA achieved 56.03% accuracy, surpassing LoRA (42.30%), AdaLoRA (47.30%), Full FT (49.05%), and PiSSA (53.07%). On the MATH benchmark, SORSA achieved 10.36% accuracy, outperforming LoRA (5.50%), AdaLoRA (6.48%), Full FT (7.22%), and PiSSA (7.44%). We conclude that SORSA offers a new perspective on parameter-efficient fine-tuning, demonstrating remarkable performance.
Authors: Saber Malekmohammadi, Golnoosh Farnadi
Abstract: A significant approach in natural language processing involves large-scale pre-training models on general domain data followed by their adaptation to specific tasks or domains. As models grow in size, full fine-tuning all of their parameters becomes increasingly impractical. To address this, some methods for low-rank task adaptation of language models have been proposed, e.g., LoRA and FLoRA. These methods keep the pre-trained model weights fixed and incorporate trainable low-rank decomposition matrices into some layers of the transformer architecture, called adapters. This approach significantly reduces the number of trainable parameters required for downstream tasks compared to full fine-tuning all parameters. In this work, we look at low-rank adaptation from the lens of data privacy. We show theoretically that the low-rank adaptation used in LoRA and FLoRA is equivalent to injecting some random noise into the batch gradients w.r.t the adapter parameters, and we quantify the variance of the injected noise. By establishing a Berry-Esseen type bound on the total variation distance between distribution of the injected noise and a Gaussian distribution with the same variance, we show that the dynamics of low-rank adaptation is close to that of differentially private fine-tuning of the adapters. Finally, using Johnson-Lindenstrauss lemma, we show that when augmented with gradient scaling, low-rank adaptation is very close to performing DPSGD algorithm with a fixed noise scale to fine-tune the adapters. These theoretical findings suggest that unlike other existing fine-tuning algorithms, low-rank adaptation provides privacy w.r.t the fine-tuning data implicitly.
Authors: Chengsen Wang, Qi Qi, Jingyu Wang, Haifeng Sun, Zirui Zhuang, Jinming Wu, Jianxin Liao
Abstract: Time series forecasting has played a pivotal role across various industries, including finance, transportation, energy, healthcare, and climate. Due to the abundant seasonal information they contain, timestamps possess the potential to offer robust global guidance for forecasting techniques. However, existing works primarily focus on local observations, with timestamps being treated merely as an optional supplement that remains underutilized. When data gathered from the real world is polluted, the absence of global information will damage the robust prediction capability of these algorithms. To address these problems, we propose a novel framework named GLAFF. Within this framework, the timestamps are modeled individually to capture the global dependencies. Working as a plugin, GLAFF adaptively adjusts the combined weights for global and local information, enabling seamless collaboration with any time series forecasting backbone. Extensive experiments conducted on nine real-world datasets demonstrate that GLAFF significantly enhances the average performance of widely used mainstream forecasting models by 12.5%, surpassing the previous state-of-the-art method by 5.5%.
Authors: Quyen Tran, Hoang Phan, Minh Le, Tuan Truong, Dinh Phung, Linh Ngo, Thien Nguyen, Nhat Ho, Trung Le
Abstract: Drawing inspiration from human learning behaviors, this work proposes a novel approach to mitigate catastrophic forgetting in Prompt-based Continual Learning models by exploiting the relationships between continuously emerging class data. We find that applying human habits of organizing and connecting information can serve as an efficient strategy when training deep learning models. Specifically, by building a hierarchical tree structure based on the expanding set of labels, we gain fresh insights into the data, identifying groups of similar classes could easily cause confusion. Additionally, we delve deeper into the hidden connections between classes by exploring the original pretrained model's behavior through an optimal transport-based approach. From these insights, we propose a novel regularization loss function that encourages models to focus more on challenging knowledge areas, thereby enhancing overall performance. Experimentally, our method demonstrated significant superiority over the most robust state-of-the-art models on various benchmarks.
Authors: Bo Tang, Elias B. Khalil, J\'an Drgo\v{n}a
Abstract: Mixed-integer non-linear programs (MINLPs) arise in various domains, such as energy systems and transportation, but are notoriously difficult to solve. Recent advances in machine learning have led to remarkable successes in optimization tasks, an area broadly known as learning to optimize. This approach includes using predictive models to generate solutions for optimization problems with continuous decision variables, thereby avoiding the need for computationally expensive optimization algorithms. However, applying learning to MINLPs remains challenging primarily due to the presence of integer decision variables, which complicate gradient-based learning. To address this limitation, we propose two differentiable correction layers that generate integer outputs while preserving gradient information. Combined with a soft penalty for constraint violation, our framework can tackle both the integrality and non-linear constraints in a MINLP. Experiments on three problem classes with convex/non-convex objective/constraints and integer/mixed-integer variables show that the proposed learning-based approach consistently produces high-quality solutions for parametric MINLPs extremely quickly. As problem size increases, traditional exact solvers and heuristic methods struggle to find feasible solutions, whereas our approach continues to deliver reliable results. Our work extends the scope of learning-to-optimize to MINLP, paving the way for integrating integer constraints into deep learning models. Our code is available at https://github.com/pnnl/L2O-pMINLP.
Authors: Tonghan Wang, Heng Dong, Yanchen Jiang, David C. Parkes, Milind Tambe
Abstract: Multiagent systems grapple with partial observability (PO), and the decentralized POMDP (Dec-POMDP) model highlights the fundamental nature of this challenge. Whereas recent approaches to addressing PO have appealed to deep learning models, providing a rigorous understanding of how these models and their approximation errors affect agents' handling of PO and their interactions remain a challenge. In addressing this challenge, we investigate reconstructing global states from local action-observation histories in Dec-POMDPs using diffusion models. We first find that diffusion models conditioned on local history represent possible states as stable fixed points. In collectively observable (CO) Dec-POMDPs, individual diffusion models conditioned on agents' local histories share a unique fixed point corresponding to the global state, while in non-CO settings, the shared fixed points yield a distribution of possible states given joint history. We further find that, with deep learning approximation errors, fixed points can deviate from true states and the deviation is negatively correlated to the Jacobian rank. Inspired by this low-rank property, we bound the deviation by constructing a surrogate linear regression model that approximates the local behavior of diffusion models. With this bound, we propose a composite diffusion process iterating over agents with theoretical convergence guarantees to the true state.
Authors: Robert Spencer (Hennie), Surangika Ranathunga (Hennie), Mikael Boulic (Hennie), Andries (Hennie), van Heerden, Teo Susnjak
Abstract: This study investigates the application of Transfer Learning (TL) on Transformer architectures to enhance building energy consumption forecasting. Transformers are a relatively new deep learning architecture, which has served as the foundation for groundbreaking technologies such as ChatGPT. While TL has been studied in the past, prior studies considered either one data-centric TL strategy or used older deep learning models such as Recurrent Neural Networks or Convolutional Neural Networks. Here, we carry out an extensive empirical study on six different data-centric TL strategies and analyse their performance under varying feature spaces. In addition to the vanilla Transformer architecture, we also experiment with Informer and PatchTST, specifically designed for time series forecasting. We use 16 datasets from the Building Data Genome Project 2 to create building energy consumption forecasting models. Experimental results reveal that while TL is generally beneficial, especially when the target domain has no data, careful selection of the exact TL strategy should be made to gain the maximum benefit. This decision largely depends on the feature space properties such as the recorded weather features. We also note that PatchTST outperforms the other two Transformer variants (vanilla Transformer and Informer). Our findings advance the building energy consumption forecasting using advanced approaches like TL and Transformer architectures.
Authors: Robin Janssen, Immanuel Sulzer, Tobias Buck
Abstract: We introduce CODES, a benchmark for comprehensive evaluation of surrogate architectures for coupled ODE systems. Besides standard metrics like mean squared error (MSE) and inference time, CODES provides insights into surrogate behaviour across multiple dimensions like interpolation, extrapolation, sparse data, uncertainty quantification and gradient correlation. The benchmark emphasizes usability through features such as integrated parallel training, a web-based configuration generator, and pre-implemented baseline models and datasets. Extensive documentation ensures sustainability and provides the foundation for collaborative improvement. By offering a fair and multi-faceted comparison, CODES helps researchers select the most suitable surrogate for their specific dataset and application while deepening our understanding of surrogate learning behaviour.
Authors: Sepehr Nourmohammadi, Arda Sarp Yenicesu, Shervin Rahimzadeh Arashloo, Ozgur S. Oguz
Abstract: This paper presents a novel approach to one-class classifier fusion through locally adaptive learning with dynamic $\ell$p-norm constraints. We introduce a framework that dynamically adjusts fusion weights based on local data characteristics, addressing fundamental challenges in ensemble-based anomaly detection. Our method incorporates an interior-point optimization technique that significantly improves computational efficiency compared to traditional Frank-Wolfe approaches, achieving up to 19-fold speed improvements in complex scenarios. The framework is extensively evaluated on standard UCI benchmark datasets and specialized temporal sequence datasets, demonstrating superior performance across diverse anomaly types. Statistical validation through Skillings-Mack tests confirms our method's significant advantages over existing approaches, with consistent top rankings in both pure and non-pure learning scenarios. The framework's ability to adapt to local data patterns while maintaining computational efficiency makes it particularly valuable for real-time applications where rapid and accurate anomaly detection is crucial.
Authors: Tongxin Li, Tinashe Handina, Shaolei Ren, Adam Wierman
Abstract: The combination of the Bayesian game and learning has a rich history, with the idea of controlling a single agent in a system composed of multiple agents with unknown behaviors given a set of types, each specifying a possible behavior for the other agents. The idea is to plan an agent's own actions with respect to those types which it believes are most likely to maximize the payoff. However, the type beliefs are often learned from past actions and likely to be incorrect. With this perspective in mind, we consider an agent in a game with type predictions of other components, and investigate the impact of incorrect beliefs to the agent's payoff. In particular, we formally define a tradeoff between risk and opportunity by comparing the payoff obtained against the optimal payoff, which is represented by a gap caused by trusting or distrusting the learned beliefs. Our main results characterize the tradeoff by establishing upper and lower bounds on the Pareto front for both normal-form and stochastic Bayesian games, with numerical results provided.
Authors: Fenghua Ling, Kang Chen, Jiye Wu, Tao Han, Jing-Jia Luo, Wanli Ouyang, Lei Bai
Abstract: Seamless forecasting that produces warning information at continuum timescales based on only one system is a long-standing pursuit for weather-climate service. While the rapid advancement of deep learning has induced revolutionary changes in classical forecasting field, current efforts are still focused on building separate AI models for weather and climate forecasts. To explore the seamless forecasting ability based on one AI model, we propose FengWu-Weather to Subseasonal (FengWu-W2S), which builds on the FengWu global weather forecast model and incorporates an ocean-atmosphere-land coupling structure along with a diverse perturbation strategy. FengWu-W2S can generate 6-hourly atmosphere forecasts extending up to 42 days through an autoregressive and seamless manner. Our hindcast results demonstrate that FengWu-W2S reliably predicts atmospheric conditions out to 3-6 weeks ahead, enhancing predictive capabilities for global surface air temperature, precipitation, geopotential height and intraseasonal signals such as the Madden-Julian Oscillation (MJO) and North Atlantic Oscillation (NAO). Moreover, our ablation experiments on forecast error growth from daily to seasonal timescales reveal potential pathways for developing AI-based integrated system for seamless weather-climate forecasting in the future.
Authors: Ankita Agarwal, Tanvi Banerjee, William Romine, Mia Cajita
Abstract: Artificial intelligence based predictive models trained on the clinical notes can be demographically biased. This could lead to adverse healthcare disparities in predicting outcomes like length of stay of the patients. Thus, it is necessary to mitigate the demographic biases within these models. We proposed an implicit in-processing debiasing method to combat disparate treatment which occurs when the machine learning model predict different outcomes for individuals based on the sensitive attributes like gender, ethnicity, race, and likewise. For this purpose, we used clinical notes of heart failure patients and used diagnostic codes, procedure reports and physiological vitals of the patients. We used Clinical BERT to obtain feature embeddings within the diagnostic codes and procedure reports, and LSTM autoencoders to obtain feature embeddings within the physiological vitals. Then, we trained two separate deep learning contrastive learning frameworks, one for gender and the other for ethnicity to obtain debiased representations within those demographic traits. We called this debiasing framework Debias-CLR. We leveraged clinical phenotypes of the patients identified in the diagnostic codes and procedure reports in the previous study to measure fairness statistically. We found that Debias-CLR was able to reduce the Single-Category Word Embedding Association Test (SC-WEAT) effect size score when debiasing for gender and ethnicity. We further found that to obtain fair representations in the embedding space using Debias-CLR, the accuracy of the predictive models on downstream tasks like predicting length of stay of the patients did not get reduced as compared to using the un-debiased counterparts for training the predictive models. Hence, we conclude that our proposed approach, Debias-CLR is fair and representative in mitigating demographic biases and can reduce health disparities.
Authors: Yueyang Shen, Agus Sudjianto, Arun Prakash R, Anwesha Bhattacharyya, Maorong Rao, Yaqun Wang, Joel Vaughan, Nengfeng Zhou
Abstract: We propose and study a minimalist approach towards synthetic tabular data generation. The model consists of a minimalistic unsupervised SparsePCA encoder (with contingent clustering step or log transformation to handle nonlinearity) and XGboost decoder which is SOTA for structured data regression and classification tasks. We study and contrast the methodologies with (variational) autoencoders in several toy low dimensional scenarios to derive necessary intuitions. The framework is applied to high dimensional simulated credit scoring data which parallels real-life financial applications. We applied the method to robustness testing to demonstrate practical use cases. The case study result suggests that the method provides an alternative to raw and quantile perturbation for model robustness testing. We show that the method is simplistic, guarantees interpretability all the way through, does not require extra tuning and provide unique benefits.
Authors: Mengxuan Li, Ke Liu, Hongyang Chen, Jiajun Bu, Hongwei Wang, Haishuai Wang
Abstract: Time series anomaly detection aims to identify unusual patterns in data or deviations from systems' expected behavior. The reconstruction-based methods are the mainstream in this task, which learn point-wise representation via unsupervised learning. However, the unlabeled anomaly points in training data may cause these reconstruction-based methods to learn and reconstruct anomalous data, resulting in the challenge of capturing normal patterns. In this paper, we propose a time series anomaly detection method based on implicit neural representation (INR) reconstruction, named TSINR, to address this challenge. Due to the property of spectral bias, TSINR enables prioritizing low-frequency signals and exhibiting poorer performance on high-frequency abnormal data. Specifically, we adopt INR to parameterize time series data as a continuous function and employ a transformer-based architecture to predict the INR of given data. As a result, the proposed TSINR method achieves the advantage of capturing the temporal continuity and thus is more sensitive to discontinuous anomaly data. In addition, we further design a novel form of INR continuous function to learn inter- and intra-channel information, and leverage a pre-trained large language model to amplify the intense fluctuations in anomalies. Extensive experiments demonstrate that TSINR achieves superior overall performance on both univariate and multivariate time series anomaly detection benchmarks compared to other state-of-the-art reconstruction-based methods. Our codes are available.
Authors: Stanley Wei, Sadhika Malladi, Sanjeev Arora, Amartya Sanyal
Abstract: Machine unlearning algorithms are increasingly important as legal concerns arise around the provenance of training data, but verifying the success of unlearning is often difficult. Provable guarantees for unlearning are often limited to supervised learning settings. In this paper, we provide the first theoretical guarantees for unlearning in the pre-training and fine-tuning paradigm by studying topic models, simple bag-of-words language models that can be adapted to solve downstream tasks like retrieval and classification. First, we design a provably effective unlearning algorithm for topic models that incurs a computational overhead independent of the size of the original dataset. Our analysis additionally quantifies the deletion capacity of the model -- i.e., the number of examples that can be unlearned without incurring a significant cost in model performance. Finally, we formally extend our analyses to account for adaptation to a given downstream task. In particular, we design an efficient algorithm to perform unlearning after fine-tuning the topic model via a linear head. Notably, we show that it is easier to unlearn pre-training data from models that have been fine-tuned to a particular task, and one can unlearn this data without modifying the base model.
Authors: Nadia Daoudi, Jordan Samhi, Abdoul Kader Kabore, Kevin Allix, Tegawend\'e F. Bissyand\'e, Jacques Klein
Abstract: Computer vision has witnessed several advances in recent years, with unprecedented performance provided by deep representation learning research. Image formats thus appear attractive to other fields such as malware detection, where deep learning on images alleviates the need for comprehensively hand-crafted features generalising to different malware variants. We postulate that this research direction could become the next frontier in Android malware detection, and therefore requires a clear roadmap to ensure that new approaches indeed bring novel contributions. We contribute with a first building block by developing and assessing a baseline pipeline for image-based malware detection with straightforward steps. We propose DexRay, which converts the bytecode of the app DEX files into grey-scale "vector" images and feeds them to a 1-dimensional Convolutional Neural Network model. We view DexRay as foundational due to the exceedingly basic nature of the design choices, allowing to infer what could be a minimal performance that can be obtained with image-based learning in malware detection. The performance of DexRay evaluated on over 158k apps demonstrates that, while simple, our approach is effective with a high detection rate (F1-score= 0.96). Finally, we investigate the impact of time decay and image-resizing on the performance of DexRay and assess its resilience to obfuscation. This work-in-progress paper contributes to the domain of Deep Learning based Malware detection by providing a sound, simple, yet effective approach (with available artefacts) that can be the basis to scope the many profound questions that will need to be investigated to fully develop this domain.
Authors: Sihan Zeng, Thinh T. Doan, Justin Romberg
Abstract: We consider a discounted cost constrained Markov decision process (CMDP) policy optimization problem, in which an agent seeks to maximize a discounted cumulative reward subject to a number of constraints on discounted cumulative utilities. To solve this constrained optimization program, we study an online actor-critic variant of a classic primal-dual method where the gradients of both the primal and dual functions are estimated using samples from a single trajectory generated by the underlying time-varying Markov processes. This online primal-dual natural actor-critic algorithm maintains and iteratively updates three variables: a dual variable (or Lagrangian multiplier), a primal variable (or actor), and a critic variable used to estimate the gradients of both primal and dual variables. These variables are updated simultaneously but on different time scales (using different step sizes) and they are all intertwined with each other. Our main contribution is to derive a finite-time analysis for the convergence of this algorithm to the global optimum of a CMDP problem. Specifically, we show that with a proper choice of step sizes the optimality gap and constraint violation converge to zero in expectation at a rate $\mathcal{O}(1/K^{1/6})$, where K is the number of iterations. To our knowledge, this paper is the first to study the finite-time complexity of an online primal-dual actor-critic method for solving a CMDP problem. We also validate the effectiveness of this algorithm through numerical simulations.
Authors: Razieh Nabi, Rohit Bhattacharya, Ilya Shpitser, James M. Robins
Abstract: It is often said that the fundamental problem of causal inference is a missing data problem -- the comparison of responses to two hypothetical treatment assignments is made difficult because for every experimental unit only one potential response is observed. In this paper, we consider the implications of the converse view: that missing data problems are a form of causal inference. We make explicit how the missing data problem of recovering the complete data law from the observed law can be viewed as identification of a joint distribution over counterfactual variables corresponding to values had we (possibly contrary to fact) been able to observe them. Drawing analogies with causal inference, we show how identification assumptions in missing data can be encoded in terms of graphical models defined over counterfactual and observed variables. We review recent results in missing data identification from this viewpoint. In doing so, we note interesting similarities and differences between missing data and causal identification theories.
Authors: Diego Garc\'ia-Mart\'in, Martin Larocca, M. Cerezo
Abstract: It is well known that artificial neural networks initialized from independent and identically distributed priors converge to Gaussian processes in the limit of a large number of neurons per hidden layer. In this work we prove an analogous result for Quantum Neural Networks (QNNs). Namely, we show that the outputs of certain models based on Haar random unitary or orthogonal deep QNNs converge to Gaussian processes in the limit of large Hilbert space dimension $d$. The derivation of this result is more nuanced than in the classical case due to the role played by the input states, the measurement observable, and the fact that the entries of unitary matrices are not independent. Then, we show that the efficiency of predicting measurements at the output of a QNN using Gaussian process regression depends on the observable's bodyness. Furthermore, our theorems imply that the concentration of measure phenomenon in Haar random QNNs is worse than previously thought, as we prove that expectation values and gradients concentrate as $\mathcal{O}\left(\frac{1}{e^d \sqrt{d}}\right)$. Finally, we discuss how our results improve our understanding of concentration in $t$-designs.
Authors: Liangqi Yuan, Yuan Wei, Jia Li
Abstract: With the emphasis on healthcare, early childhood education, and fitness, non-invasive measurement and recognition methods have received more attention. Pressure sensing has been extensively studied because of its advantages of simple structure, easy access, visualization application, and harmlessness. This paper introduces a Smart Pressure e-Mat (SPeM) system based on piezoresistive material, Velostat, for human monitoring applications, including recognition of sleeping postures, sports, and yoga. After a subsystem scans the e-mat readings and processes the signal, it generates a pressure image stream. Deep neural networks (DNNs) are used to fit and train the pressure image stream and recognize the corresponding human behavior. Four sleeping postures and 13 dynamic activities inspired by Nintendo Switch Ring Fit Adventure (RFA) are used as a preliminary validation of the proposed SPeM system. The SPeM system achieves high accuracies in both applications, demonstrating the high accuracy and generalizability of the models. Compared with other pressure sensor-based systems, SPeM possesses more flexible applications and commercial application prospects, with reliable, robust, and repeatable properties.
Authors: Mehdi Letafati, Samad Ali, Matti Latva-aho
Abstract: In this paper, conditional denoising diffusion probabilistic models (DDPMs) are proposed to enhance the data transmission and reconstruction over wireless channels. The underlying mechanism of DDPM is to decompose the data generation process over the so-called "denoising" steps. Inspired by this, the key idea is to leverage the generative prior of diffusion models in learning a "noisy-to-clean" transformation of the information signal to help enhance data reconstruction. The proposed scheme could be beneficial for communication scenarios in which a prior knowledge of the information content is available, e.g., in multimedia transmission. Hence, instead of employing complicated channel codes that reduce the information rate, one can exploit diffusion priors for reliable data reconstruction, especially under extreme channel conditions due to low signal-to-noise ratio (SNR), or hardware-impaired communications. The proposed DDPM-assisted receiver is tailored for the scenario of wireless image transmission using MNIST dataset. Our numerical results highlight the reconstruction performance of our scheme compared to the conventional digital communication, as well as the deep neural network (DNN)-based benchmark. It is also shown that more than 10 dB improvement in the reconstruction could be achieved in low SNR regimes, without the need to reduce the information rate for error correction.
Authors: Hongbin Ye, Honghao Gui, Aijia Zhang, Tong Liu, Weiqiang Jia
Abstract: This paper introduces CooperKGC, a novel framework challenging the conventional solitary approach of large language models (LLMs) in knowledge graph construction (KGC). CooperKGC establishes a collaborative processing network, assembling a team capable of concurrently addressing entity, relation, and event extraction tasks. Experimentation demonstrates that fostering collaboration within CooperKGC enhances knowledge selection, correction, and aggregation capabilities across multiple rounds of interactions.
Authors: Jian Gu, Aldeida Aleti, Chunyang Chen, Hongyu Zhang
Abstract: Language Models (LMs) have become widely used in software engineering, especially for tasks such as code generation, where they are referred to as code LMs. These models have proven effective in generating code, making it easier for developers to automate coding activities. However, research has highlighted a significant limitation: despite their effectiveness, LMs often produce code that is incorrect, buggy, or not fully functional. Updating these models with limited data can be prohibitively challenging, yet it is essential to maximize their utility. This may require hot-fix techniques (updating models with limited data) to resolve. In this paper, we propose \ul{M}odel \ul{I}mprovement via \ul{N}euron \ul{T}argeting (\textsc{MINT}), a novel approach for repairing code LMs. MINT leverages the semantic property of language models to perform neuron-level repairs in a novel way. Further, by analyzing the relationships between the model's latent representations, the incorrect outputs, and the desired outputs, \textsc{MINT} determines which neurons are worth updating. This approach ensures that only the neurons crucial to the model's failure are targeted, avoiding unnecessary changes and allowing for a more efficient and precise repair process. \textsc{MINT} is effective, efficient, and reliable, capable of correcting a neural model by patching a minimum number of neurons (usually one or two neurons). Our approach is evaluated on three coding tasks: line-level code generation, shellcode generation, and intent-to-bash translation. The experimental results demonstrate that the proposed approach significantly outperforms the state-of-the-art in both effectiveness and efficiency measures. In addition, we analyze and discuss the side effects of model repair techniques, including the balance between generalization and specificity, and the performance after multiple repairs in succession.
Authors: Jiahong Fu, Qi Xie, Deyu Meng, Zongben Xu
Abstract: The deep unfolding approach has attracted significant attention in computer vision tasks, which well connects conventional image processing modeling manners with more recent deep learning techniques. Specifically, by establishing a direct correspondence between algorithm operators at each implementation step and network modules within each layer, one can rationally construct an almost ``white box'' network architecture with high interpretability. In this architecture, only the predefined component of the proximal operator, known as a proximal network, needs manual configuration, enabling the network to automatically extract intrinsic image priors in a data-driven manner. In current deep unfolding methods, such a proximal network is generally designed as a CNN architecture, whose necessity has been proven by a recent theory. That is, CNN structure substantially delivers the translational invariant image prior, which is the most universally possessed structural prior across various types of images. However, standard CNN-based proximal networks have essential limitations in capturing the rotation symmetry prior, another universal structural prior underlying general images. This leaves a large room for further performance improvement in deep unfolding approaches. To address this issue, this study makes efforts to suggest a high-accuracy rotation equivariant proximal network that effectively embeds rotation symmetry priors into the deep unfolding framework. Especially, we deduce, for the first time, the theoretical equivariant error for such a designed proximal network with arbitrary layers under arbitrary rotation degrees. This analysis should be the most refined theoretical conclusion for such error evaluation to date and is also indispensable for supporting the rationale behind such networks with intrinsic interpretability requirements.
Authors: Nicolas Boull\'e, Diana Halikias, Samuel E. Otto, Alex Townsend
Abstract: There is a mystery at the heart of operator learning: how can one recover a non-self-adjoint operator from data without probing the adjoint? Current practical approaches suggest that one can accurately recover an operator while only using data generated by the forward action of the operator without access to the adjoint. However, naively, it seems essential to sample the action of the adjoint. In this paper, we partially explain this mystery by proving that without querying the adjoint, one can approximate a family of non-self-adjoint infinite-dimensional compact operators via projection onto a Fourier basis. We then apply the result to recovering Green's functions of elliptic partial differential operators and derive an adjoint-free sample complexity bound. While existing theory justifies low sample complexity in operator learning, ours is the first adjoint-free analysis that attempts to close the gap between theory and practice.
Authors: Andrew Ye, James Xu, Vidyut Veedgav, Yi Wang, Yifan Yu, Daniel Yan, Ryan Chen, Vipin Chaudhary, Shuai Xu
Abstract: We propose and study the integration of sentiment analysis and deep reinforcement learning ensemble algorithms for stock trading by evaluating strategies capable of dynamically altering their active agent given the concurrent market environment. In particular, we design a simple-yet-effective method for extracting financial sentiment and combine this with improvements on existing trading agents, resulting in a strategy that effectively considers both qualitative market factors and quantitative stock data. We show that our approach results in a strategy that is profitable, robust, and risk-minimal - outperforming the traditional ensemble strategy as well as single agent algorithms and market metrics. Our findings suggest that the conventional practice of switching and reevaluating agents in ensemble every fixed-number of months is sub-optimal, and that a dynamic sentiment-based framework greatly unlocks additional performance. Furthermore, as we have designed our algorithm with simplicity and efficiency in mind, we hypothesize that the transition of our method from historical evaluation towards real-time trading with live data to be relatively simple.
Authors: Ameya Prabhu, Shiven Sinha, Ponnurangam Kumaraguru, Philip H. S. Torr, Ozan Sener, Puneet K. Dokania
Abstract: Continual learning has primarily focused on the issue of catastrophic forgetting and the associated stability-plasticity tradeoffs. However, little attention has been paid to the efficacy of continually learned representations, as representations are learned alongside classifiers throughout the learning process. Our primary contribution is empirically demonstrating that existing online continually trained deep networks produce inferior representations compared to a simple pre-defined random transforms. Our approach projects raw pixels using a fixed random transform, approximating an RBF-Kernel initialized before any data is seen. We then train a simple linear classifier on top without storing any exemplars, processing one sample at a time in an online continual learning setting. This method, called RanDumb, significantly outperforms state-of-the-art continually learned representations across all standard online continual learning benchmarks. Our study reveals the significant limitations of representation learning, particularly in low-exemplar and online continual learning scenarios. Extending our investigation to popular exemplar-free scenarios with pretrained models, we find that training only a linear classifier on top of pretrained representations surpasses most continual fine-tuning and prompt-tuning strategies. Overall, our investigation challenges the prevailing assumptions about effective representation learning in online continual learning. Our code is available at://github.com/drimpossible/RanDumb.
Authors: Rahul Mishra, Arnav Bhavsar
Abstract: In this study, we tackle a modern research challenge within the field of perceptual brain decoding, which revolves around synthesizing images from EEG signals using an adversarial deep learning framework. The specific objective is to recreate images belonging to various object categories by leveraging EEG recordings obtained while subjects view those images. To achieve this, we employ a Transformer-encoder based EEG encoder to produce EEG encodings, which serve as inputs to the generator component of the GAN network. Alongside the adversarial loss, we also incorporate perceptual loss to enhance the quality of the generated images.
Authors: Xuechen Zhang, Zijian Huang, Ege Onur Taga, Carlee Joe-Wong, Samet Oymak, Jiasi Chen
Abstract: Recent successes in natural language processing have led to the proliferation of large language models (LLMs) by multiple providers. Each LLM offering has different inference accuracy, monetary cost, and latency, and their accuracy further depends on the exact wording of the question (i.e., the specific prompt). At the same time, users often have a limit on monetary budget and latency to answer all their questions, and they do not know which LLMs to choose for each question to meet their accuracy and long term budget requirements. To navigate this rich design space, we propose TREACLE ($\underline{T}$hrifty $\underline{Rea}$soning via $\underline{C}$ontext-Aware $\underline{L}$LM and Prompt S$\underline{e}$lection), a reinforcement learning policy that jointly selects the model and prompting scheme while respecting the user's monetary cost and latency constraints. TREACLE uses the problem context, including question text embeddings (reflecting the type or difficulty of a query) and the response history (reflecting the consistency of previous responses) to make smart decisions. Our evaluations on standard reasoning datasets (GSM8K, CSQA, and LLC) with various LLMs and prompts show that TREACLE enables cost savings of up to 85% compared to baselines, while maintaining high accuracy. Importantly, it provides the user with the ability to gracefully trade off accuracy for cost.
Authors: Nicholas Clinton, Andreas Vollrath, Remi D'annunzio, Desheng Liu, Henry B. Glick, Adri\`a Descals, Alicia Sullivan, Oliver Guinan, Jacob Abramowitz, Fred Stolle, Chris Goodman, Tanya Birch, David Quinn, Olga Danylo, Tijs Lips, Daniel Coelho, Enikoe Bihari, Bryce Cronkite-Ratcliff, Ate Poortinga, Atena Haghighattalab, Evan Notman, Michael DeWitt, Aaron Yonas, Gennadii Donchyts, Devaja Shah, David Saah, Karis Tenneson, Nguyen Hanh Quyen, Megha Verma, Andrew Wilcox
Abstract: Palm oil production has been identified as one of the major drivers of deforestation for tropical countries. To meet supply chain objectives, commodity producers and other stakeholders need timely information of land cover dynamics in their supply shed. However, such data are difficult to obtain from suppliers who may lack digital geographic representations of their supply sheds and production locations. Here we present a "community model," a machine learning model trained on pooled data sourced from many different stakeholders, to produce a map of palm probability at global scale. An advantage of this method is the inclusion of varied inputs, the ability to easily update the model as new training data becomes available and run the model on any year that input imagery is available. Inclusion of diverse data sources into one probability map can help establish a shared understanding across stakeholders on the presence and absence of a land cover or commodity (in this case oil palm). The model predictors are annual composites built from publicly available satellite imagery provided by Sentinel-1, Sentinel-2, and ALOS-2, and terrain data from Jaxa (AW3D30) and Copernicus (GLO-30). We provide map outputs as the probability of palm in a given pixel, to reflect the uncertainty of the underlying state (palm or not palm). This version of this model provides global accuracy estimated to be 92% (at 0.5 probability threshold) on an independent test set. This model, and resulting oil palm probability map products are useful for accurately identifying the geographic footprint of palm cultivation. Used in conjunction with timely deforestation information, this palm model is useful for understanding the risk of continued oil palm plantation expansion in sensitive forest areas.
Authors: Franck Iutzeler, Edouard Pauwels, Samuel Vaiter
Abstract: We consider stochastic optimization problems where the objective depends on some parameter, as commonly found in hyperparameter optimization for instance. We investigate the behavior of the derivatives of the iterates of Stochastic Gradient Descent (SGD) with respect to that parameter and show that they are driven by an inexact SGD recursion on a different objective function, perturbed by the convergence of the original SGD. This enables us to establish that the derivatives of SGD converge to the derivative of the solution mapping in terms of mean squared error whenever the objective is strongly convex. Specifically, we demonstrate that with constant step-sizes, these derivatives stabilize within a noise ball centered at the solution derivative, and that with vanishing step-sizes they exhibit $O(\log(k)^2 / k)$ convergence rates. Additionally, we prove exponential convergence in the interpolation regime. Our theoretical findings are illustrated by numerical experiments on synthetic tasks.
Authors: Sean Welleck, Amanda Bertsch, Matthew Finlayson, Hailey Schoelkopf, Alex Xie, Graham Neubig, Ilia Kulikov, Zaid Harchaoui
Abstract: One of the most striking findings in modern research on large language models (LLMs) is that scaling up compute during training leads to better results. However, less attention has been given to the benefits of scaling compute during inference. This survey focuses on these inference-time approaches. We explore three areas under a unified mathematical formalism: token-level generation algorithms, meta-generation algorithms, and efficient generation. Token-level generation algorithms, often called decoding algorithms, operate by sampling a single token at a time or constructing a token-level search space and then selecting an output. These methods typically assume access to a language model's logits, next-token distributions, or probability scores. Meta-generation algorithms work on partial or full sequences, incorporating domain knowledge, enabling backtracking, and integrating external information. Efficient generation methods aim to reduce token costs and improve the speed of generation. Our survey unifies perspectives from three research communities: traditional natural language processing, modern LLMs, and machine learning systems.
Authors: Bailu Ding, Jiaqi Zhai
Abstract: Retrieval plays a fundamental role in recommendation systems, search, and natural language processing (NLP) by efficiently finding relevant items from a large corpus given a query. Dot products have been widely used as the similarity function in such tasks, enabled by Maximum Inner Product Search (MIPS) algorithms for efficient retrieval. However, state-of-the-art retrieval algorithms have migrated to learned similarities. These advanced approaches encompass multiple query embeddings, complex neural networks, direct item ID decoding via beam search, and hybrid solutions. Unfortunately, we lack efficient solutions for retrieval in these state-of-the-art setups. Our work addresses this gap by investigating efficient retrieval techniques with expressive learned similarity functions. We establish Mixture-of-Logits (MoL) as a universal approximator of similarity functions, demonstrate that MoL's expressiveness can be realized empirically to achieve superior performance on diverse retrieval scenarios, and propose techniques to retrieve the approximate top-k results using MoL with tight error bounds. Through extensive experimentation, we show that MoL, enhanced by our proposed mutual information-based load balancing loss, sets new state-of-the-art results across heterogeneous scenarios, including sequential retrieval models in recommendation systems and finetuning language models for question answering; and our approximate top-$k$ algorithms outperform baselines by up to 66x in latency while achieving >.99 recall rate compared to exact algorithms.
Authors: Nivasini Ananthakrishnan, Nika Haghtalab, Chara Podimata, Kunhe Yang
Abstract: When learning in strategic environments, a key question is whether agents can overcome uncertainty about their preferences to achieve outcomes they could have achieved absent any uncertainty. Can they do this solely through interactions with each other? We focus this question on the ability of agents to attain the value of their Stackelberg optimal strategy and study the impact of information asymmetry. We study repeated interactions in fully strategic environments where players' actions are decided based on learning algorithms that take into account their observed histories and knowledge of the game. We study the pure Nash equilibria (PNE) of a meta-game where players choose these algorithms as their actions. We demonstrate that if one player has perfect knowledge about the game, then any initial informational gap persists. That is, while there is always a PNE in which the informed agent achieves her Stackelberg value, there is a game where no PNE of the meta-game allows the partially informed player to achieve her Stackelberg value. On the other hand, if both players start with some uncertainty about the game, the quality of information alone does not determine which agent can achieve her Stackelberg value. In this case, the concept of information asymmetry becomes nuanced and depends on the game's structure. Overall, our findings suggest that repeated strategic interactions alone cannot facilitate learning effectively enough to earn an uninformed player her Stackelberg value.
Authors: Yash Bhalgat, Vadim Tschernezki, Iro Laina, Jo\~ao F. Henriques, Andrea Vedaldi, Andrew Zisserman
Abstract: Egocentric videos present unique challenges for 3D scene understanding due to rapid camera motion, frequent object occlusions, and limited object visibility. This paper introduces a novel approach to instance segmentation and tracking in first-person video that leverages 3D awareness to overcome these obstacles. Our method integrates scene geometry, 3D object centroid tracking, and instance segmentation to create a robust framework for analyzing dynamic egocentric scenes. By incorporating spatial and temporal cues, we achieve superior performance compared to state-of-the-art 2D approaches. Extensive evaluations on the challenging EPIC Fields dataset demonstrate significant improvements across a range of tracking and segmentation consistency metrics. Specifically, our method outperforms the next best performing approach by $7$ points in Association Accuracy (AssA) and $4.5$ points in IDF1 score, while reducing the number of ID switches by $73\%$ to $80\%$ across various object categories. Leveraging our tracked instance segmentations, we showcase downstream applications in 3D object reconstruction and amodal video object segmentation in these egocentric settings.
Authors: Jithu J Athalathil, Bhargav Vaidya, Sayan Kundu, Vishal Upendran, Mark C. M. Cheung
Abstract: Studying the magnetic field properties on the solar surface is crucial for understanding the solar and heliospheric activities, which in turn shape space weather in the solar system. Surface Flux Transport (SFT) modeling helps us to simulate and analyse the transport and evolution of magnetic flux on the solar surface, providing valuable insights into the mechanisms responsible for solar activity. In this work, we demonstrate the use of machine learning techniques in solving magnetic flux transport, making it accurate. We have developed a novel Physics-Informed Neural Networks (PINN)-based model to study the evolution of Bipolar Magnetic Regions (BMRs) using SFT in one-dimensional azimuthally averaged and also in two-dimensions. We demonstrate the efficiency and computational feasibility of our PINN-based model by comparing its performance and accuracy with that of a numerical model implemented using the Runge-Kutta Implicit-Explicit (RK-IMEX) scheme. The mesh-independent PINN method can be used to reproduce the observed polar magnetic field with better flux conservation. This advancement is important for accurately reproducing observed polar magnetic fields, thereby providing insights into the strength of future solar cycles. This work paves the way for more efficient and accurate simulations of solar magnetic flux transport and showcases the applicability of PINN in solving advection-diffusion equations with a particular focus on heliophysics.
Authors: Alexis Vieloszynski, Soumaya Cherkaoui, Ola Ahmad, Jean-Fr\'ed\'eric Laprade, Oliver Nahman-L\'evesque, Abdallah Aaraba, Shengrui Wang
Abstract: Quantum machine learning consists in taking advantage of quantum computations to generate classical data. A potential application of quantum machine learning is to harness the power of quantum computers for generating classical data, a process essential to a multitude of applications such as enriching training datasets, anomaly detection, and risk management in finance. Given the success of Generative Adversarial Networks in classical image generation, the development of its quantum versions has been actively conducted. However, existing implementations on quantum computers often face significant challenges, such as scalability and training convergence issues. To address these issues, we propose LatentQGAN, a novel quantum model that uses a hybrid quantum-classical GAN coupled with an autoencoder. Although it was initially designed for image generation, the LatentQGAN approach holds potential for broader application across various practical data generation tasks. Experimental outcomes on both classical simulators and noisy intermediate scale quantum computers have demonstrated significant performance enhancements over existing quantum methods, alongside a significant reduction in quantum resources overhead.
Authors: Douba Jafuno, Ammar Mian, Guillaume Ginolhac, Nickolas Stelzenmuller
Abstract: In this paper, a new classification model based on covariance matrices is built in order to classify buried objects. The inputs of the proposed models are the hyperbola thumbnails obtained with a classical Ground Penetrating Radar (GPR) system. These thumbnails are then inputs to the first layers of a classical CNN, which then produces a covariance matrix using the outputs of the convolutional filters. Next, the covariance matrix is given to a network composed of specific layers to classify Symmetric Positive Definite (SPD) matrices. We show in a large database that our approach outperform shallow networks designed for GPR data and conventional CNNs typically used in computer vision applications, particularly when the number of training data decreases and in the presence of mislabeled data. We also illustrate the interest of our models when training data and test sets are obtained from different weather modes or considerations.
Authors: Zhumazhan Balapanov, Vanessa Matvei, Olivia Holmberg, Edward Magongo, Jonathan Pei, Kevin Zhu
Abstract: Convolutional neural networks (CNNs) have made significant advances in computer vision tasks, yet their high inference times and latency often limit real-world applicability. While model compression techniques have gained popularity as solutions, they often overlook the critical balance between low latency and uncompromised accuracy. By harnessing quantum-inspired pruning, tensor decomposition, and annealing-based matrix factorization - three quantum-inspired concepts - we introduce QIANets: a novel approach of redesigning the traditional GoogLeNet, DenseNet, and ResNet-18 model architectures to process more parameters and computations whilst maintaining low inference times. Despite experimental limitations, the method was tested and evaluated, demonstrating reductions in inference times, along with effective accuracy preservations.
Authors: Daniel Holmberg, Emanuela Clementi, Teemu Roos
Abstract: Accurate ocean forecasting systems are vital for understanding marine dynamics, which play a crucial role in environmental management and climate adaptation strategies. Traditional numerical solvers, while effective, are computationally expensive and time-consuming. Recent advancements in machine learning have revolutionized weather forecasting, offering fast and energy-efficient alternatives. Building on these advancements, we introduce SeaCast, a neural network designed for high-resolution, medium-range ocean forecasting. SeaCast employs a graph-based framework to effectively handle the complex geometry of ocean grids and integrates external forcing data tailored to the regional ocean context. Our approach is validated through experiments at a high spatial resolution using the operational numerical model of the Mediterranean Sea provided by the Copernicus Marine Service, along with both numerical and data-driven atmospheric forcings.
Authors: Mingyuan Zhou, Huangjie Zheng, Yi Gu, Zhendong Wang, Hai Huang
Abstract: Score identity Distillation (SiD) is a data-free method that has achieved SOTA performance in image generation by leveraging only a pretrained diffusion model, without requiring any training data. However, its ultimate performance is constrained by how accurate the pretrained model captures the true data scores at different stages of the diffusion process. In this paper, we introduce SiDA (SiD with Adversarial Loss), which not only enhances generation quality but also improves distillation efficiency by incorporating real images and adversarial loss. SiDA utilizes the encoder from the generator's score network as a discriminator, boosting its ability to distinguish between real images and those generated by SiD. The adversarial loss is batch-normalized within each GPU and then combined with the original SiD loss. This integration effectively incorporates the average "fakeness" per GPU batch into the pixel-based SiD loss, enabling SiDA to distill a single-step generator either from scratch or by fine-tuning an existing one. SiDA converges significantly faster than its predecessor when trained from scratch, and swiftly improves upon the original model's performance after an initial warmup period during fine-tuning from a pre-distilled SiD generator. This one-step adversarial distillation method establishes new benchmarks in generation performance when distilling EDM diffusion models pretrained on CIFAR-10 (32x32) and ImageNet (64x64), achieving FID score of 1.110 on ImageNet 64x64. It sets record-low FID scores when distilling EDM2 models trained on ImageNet (512x512), surpassing even the largest teacher model, EDM2-XXL. Our SiDA's results record FID scores of 2.156 for EDM2-XS, 1.669 for S, 1.488 for M, 1.413 for L, 1.379 for XL, and 1.366 for XXL, demonstrating significant improvements across all model sizes. Our open-source code will be integrated into the SiD codebase.
Authors: Xun Jiang, Feng Li, Han Zhao, Jiaying Wang, Jun Shao, Shihao Xu, Shu Zhang, Weiling Chen, Xavier Tang, Yize Chen, Mengyue Wu, Weizhi Ma, Mengdi Wang, Tianqiao Chen
Abstract: Large language models (LLMs) like GPTs, trained on vast datasets, have demonstrated impressive capabilities in language understanding, reasoning, and planning, achieving human-level performance in various tasks. Most studies focus on enhancing these models by training on ever-larger datasets to build more powerful foundation models. While training stronger models is important, enabling models to evolve during inference is equally crucial, a process we refer to as AI self-evolution. Unlike large-scale training, self-evolution may rely on limited data or interactions. Inspired by the columnar organization of the human cerebral cortex, we hypothesize that AI models could develop cognitive abilities and build internal representations through iterative interactions with their environment. To achieve this, models need long-term memory (LTM) to store and manage processed interaction data. LTM supports self-evolution by representing diverse experiences across environments and agents. In this report, we explore AI self-evolution and its potential to enhance models during inference. We examine LTM's role in lifelong learning, allowing models to evolve based on accumulated interactions. We outline the structure of LTM and the systems needed for effective data retention and representation. We also classify approaches for building personalized models with LTM data and show how these models achieve self-evolution through interaction. Using LTM, our multi-agent framework OMNE achieved first place on the GAIA benchmark, demonstrating LTM's potential for AI self-evolution. Finally, we present a roadmap for future research, emphasizing the importance of LTM for advancing AI technology and its practical applications.
Authors: Haokun Liu, Yangqiaoyu Zhou, Mingxuan Li, Chenfei Yuan, Chenhao Tan
Abstract: AI holds promise for transforming scientific processes, including hypothesis generation. Prior work on hypothesis generation can be broadly categorized into theory-driven and data-driven approaches. While both have proven effective in generating novel and plausible hypotheses, it remains an open question whether they can complement each other. To address this, we develop the first method that combines literature-based insights with data to perform LLM-powered hypothesis generation. We apply our method on five different datasets and demonstrate that integrating literature and data outperforms other baselines (8.97\% over few-shot, 15.75\% over literature-based alone, and 3.37\% over data-driven alone). Additionally, we conduct the first human evaluation to assess the utility of LLM-generated hypotheses in assisting human decision-making on two challenging tasks: deception detection and AI generated content detection. Our results show that human accuracy improves significantly by 7.44\% and 14.19\% on these tasks, respectively. These findings suggest that integrating literature-based and data-driven approaches provides a comprehensive and nuanced framework for hypothesis generation and could open new avenues for scientific inquiry.
Authors: Oktay Agcaoglu (for the Alzheimers Disease Neuroimaging Initiative), Rogers F. Silva (for the Alzheimers Disease Neuroimaging Initiative), Deniz Alacam (for the Alzheimers Disease Neuroimaging Initiative), Sergey Plis (for the Alzheimers Disease Neuroimaging Initiative), Tulay Adali (for the Alzheimers Disease Neuroimaging Initiative), Vince Calhoun (for the Alzheimers Disease Neuroimaging Initiative)
Abstract: Different brain imaging modalities offer unique insights into brain function and structure. Combining them enhances our understanding of neural mechanisms. Prior multimodal studies fusing functional MRI (fMRI) and structural MRI (sMRI) have shown the benefits of this approach. Since sMRI lacks temporal data, existing fusion methods often compress fMRI temporal information into summary measures, sacrificing rich temporal dynamics. Motivated by the observation that covarying networks are identified in both sMRI and resting-state fMRI, we developed a novel fusion method, by combining deep learning frameworks, copulas and independent component analysis (ICA), named copula linked parallel ICA (CLiP-ICA). This method estimates independent sources for each modality and links the spatial sources of fMRI and sMRI using a copula-based model for more flexible integration of temporal and spatial data. We tested CLiP-ICA using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Our results showed that CLiP-ICA effectively captures both strongly and weakly linked sMRI and fMRI networks, including the cerebellum, sensorimotor, visual, cognitive control, and default mode networks. It revealed more meaningful components and fewer artifacts, addressing the long-standing issue of optimal model order in ICA. CLiP-ICA also detected complex functional connectivity patterns across stages of cognitive decline, with cognitively normal subjects generally showing higher connectivity in sensorimotor and visual networks compared to patients with Alzheimer, along with patterns suggesting potential compensatory mechanisms.
Authors: Alireza Mohamadi, Hosna Ghahramani, Seyyed Amir Asghari, Mehdi Aminian
Abstract: The increasing integration of the Internet of Medical Things (IoMT) into healthcare systems has significantly enhanced patient care but has also introduced critical cybersecurity challenges. This paper presents a novel approach based on Convolutional Neural Networks (CNNs) for detecting cyberattacks within IoMT environments. Unlike previous studies that predominantly utilized traditional machine learning (ML) models or simpler Deep Neural Networks (DNNs), the proposed model leverages the capabilities of CNNs to effectively analyze the temporal characteristics of network traffic data. Trained and evaluated on the CICIoMT2024 dataset, which comprises 18 distinct types of cyberattacks across a range of IoMT devices, the proposed CNN model demonstrates superior performance compared to previous state-of-the-art methods, achieving a perfect accuracy of 99% in binary, categorical, and multiclass classification tasks. This performance surpasses that of conventional ML models such as Logistic Regression, AdaBoost, DNNs, and Random Forests. These findings highlight the potential of CNNs to substantially improve IoMT cybersecurity, thereby ensuring the protection and integrity of connected healthcare systems.
Authors: Xi Zheng, Xiangyu Chang, Ruoxi Jia, Yong Tan
Abstract: As data emerges as a vital driver of technological and economic advancements, a key challenge is accurately quantifying its value in algorithmic decision-making. The Shapley value, a well-established concept from cooperative game theory, has been widely adopted to assess the contribution of individual data sources in supervised machine learning. However, its symmetry axiom assumes all players in the cooperative game are homogeneous, which overlooks the complex structures and dependencies present in real-world datasets. To address this limitation, we extend the traditional data Shapley framework to asymmetric data Shapley, making it flexible enough to incorporate inherent structures within the datasets for structure-aware data valuation. We also introduce an efficient $k$-nearest neighbor-based algorithm for its exact computation. We demonstrate the practical applicability of our framework across various machine learning tasks and data market contexts. The code is available at: https://github.com/xzheng01/Asymmetric-Data-Shapley.
Authors: Xunze Liu, Yifei Sun, Zhaorui Wang, Lizhao You, Haoyuan Pan, Fangxin Wang, Shuguang Cui
Abstract: This paper investigates semantic communications between a transmitter and a receiver, where original data, such as videos of interest to the receiver, is stored at the transmitter. Although significant process has been made in semantic communications, a fundamental design problem is that the semantic information is extracted based on certain criteria at the transmitter alone, without considering the receiver's specific information needs. As a result, critical information of primary concern to the receiver may be lost. In such cases, the semantic transmission becomes meaningless to the receiver, as all received information is irrelevant to its interests. To solve this problem, this paper presents a receiver-centric generative semantic communication system, where each transmission is initialized by the receiver. Specifically, the receiver first sends its request for the desired semantic information to the transmitter at the start of each transmission. Then, the transmitter extracts the required semantic information accordingly. A key challenge is how the transmitter understands the receiver's requests for semantic information and extracts the required semantic information in a reasonable and robust manner. We address this challenge by designing a well-structured framework and leveraging off-the-shelf generative AI products, such as GPT-4, along with several specialized tools for detection and estimation. Evaluation results demonstrate the feasibility and effectiveness of the proposed new semantic communication system.
Authors: Daniel P. Jeong, Saurabh Garg, Zachary C. Lipton, Michael Oberst
Abstract: Several recent works seek to develop foundation models specifically for medical applications, adapting general-purpose large language models (LLMs) and vision-language models (VLMs) via continued pretraining on publicly available biomedical corpora. These works typically claim that such domain-adaptive pretraining (DAPT) improves performance on downstream medical tasks, such as answering medical licensing exam questions. In this paper, we compare seven public "medical" LLMs and two VLMs against their corresponding base models, arriving at a different conclusion: all medical VLMs and nearly all medical LLMs fail to consistently improve over their base models in the zero-/few-shot prompting regime for medical question-answering (QA) tasks. For instance, across the tasks and model pairs we consider in the 3-shot setting, medical LLMs only outperform their base models in 12.1% of cases, reach a (statistical) tie in 49.8% of cases, and are significantly worse than their base models in the remaining 38.2% of cases. Our conclusions are based on (i) comparing each medical model head-to-head, directly against the corresponding base model; (ii) optimizing the prompts for each model separately; and (iii) accounting for statistical uncertainty in comparisons. While these basic practices are not consistently adopted in the literature, our ablations show that they substantially impact conclusions. Our findings suggest that state-of-the-art general-domain models may already exhibit strong medical knowledge and reasoning capabilities, and offer recommendations to strengthen the conclusions of future studies.
Authors: Hadi Beik Mohammadi, S{\o}ren Hauberg, Georgios Arvanitidis, Gerhard Neumann, Leonel Rozo
Abstract: Stability guarantees are crucial when ensuring that a fully autonomous robot does not take undesirable or potentially harmful actions. We recently proposed the Neural Contractive Dynamical Systems (NCDS), which is a neural network architecture that guarantees contractive stability. With this, learning-from-demonstrations approaches can trivially provide stability guarantees. However, our early work left several unanswered questions, which we here address. Beyond providing an in-depth explanation of NCDS, this paper extends the framework with more careful regularization, a conditional variant of the framework for handling multiple tasks, and an uncertainty-driven approach to latent obstacle avoidance. Experiments verify that the developed system has the flexibility of ordinary neural networks while providing the stability guarantees needed for autonomous robotics.
Authors: Malte Luttermann, Ralf M\"oller, Marcel Gehrke
Abstract: Lifted probabilistic inference exploits symmetries in a probabilistic model to allow for tractable probabilistic inference with respect to domain sizes of logical variables. We found that the current state-of-the-art algorithm to construct a lifted representation in form of a parametric factor graph misses symmetries between factors that are exchangeable but scaled differently, thereby leading to a less compact representation. In this paper, we propose a generalisation of the advanced colour passing (ACP) algorithm, which is the state of the art to construct a parametric factor graph. Our proposed algorithm allows for potentials of factors to be scaled arbitrarily and efficiently detects more symmetries than the original ACP algorithm. By detecting strictly more symmetries than ACP, our algorithm significantly reduces online query times for probabilistic inference when the resulting model is applied, which we also confirm in our experiments.
Authors: Lu Wang-N\"oth, Philipp Heiler, Hai Huang, Daniel Lichtenstern, Alexandra Reichenbach, Luis Flacke, Linus Maisch, Helmut Mayer
Abstract: Objective. Electroencephalography (EEG) is a widely used neuroimaging technique known for its cost-effectiveness and user-friendliness. However, various artifacts, particularly biological artifacts like Electromyography (EMG) signals, lead to a poor signal-to-noise ratio, limiting the precision of analyses and applications. The currently reported EEG data cleaning performance largely depends on the data used for validation, and in the case of machine learning approaches, also on the data used for training. The data are typically gathered either by recruiting subjects to perform specific artifact tasks or by integrating existing datasets. Prevailing approaches, however, tend to rely on intuitive, concept-oriented data collection with minimal justification for the selection of artifacts and their quantities. Given the substantial costs associated with biological data collection and the pressing need for effective data utilization, we propose an optimization procedure for data-oriented data collection design using deep learning-based artifact detection. Approach. We apply a binary classification between artifact epochs (time intervals containing artifacts) and non-artifact epochs (time intervals containing no artifact) using three different neural architectures. Our aim is to minimize data collection efforts while preserving the cleaning efficiency. Main results. We were able to reduce the number of artifact tasks from twelve to three and decrease repetitions of isometric contraction tasks from ten to three or sometimes even just one. Significance. Our work addresses the need for effective data utilization in biological data collection, offering a systematic and dynamic quantitative approach. By providing clear justifications for the choices of artifacts and their quantity, we aim to guide future studies toward more effective and economical data collection in EEG and EMG research.
Authors: Jai Doshi, Asa Cooper Stickland
Abstract: Large language model unlearning aims to remove harmful information that LLMs have learnt to prevent their use for malicious purposes. LLMU and RMU have been proposed as two methods for LLM unlearning, achieving impressive results on unlearning benchmarks. We study in detail the efficacy of these methods by evaluating their impact on general model capabilities on the WMDP benchmark as well as a biology benchmark we create. Our experiments show that RMU generally leads to better preservation of model capabilities, for similar or better unlearning. We further test the robustness of these methods and find that doing 5-shot prompting or rephrasing the question in simple ways can lead to an over ten-fold increase in accuracy on unlearning benchmarks. Finally, we show that training on unrelated data can almost completely recover pre-unlearning performance, demonstrating that these methods fail at truly unlearning. The code is available at: https://github.com/JaiDoshi/Knowledge-Erasure.
Authors: Daeyong Kwon, SeungHeon Doh, Juhan Nam
Abstract: Intent classification is a text understanding task that identifies user needs from input text queries. While intent classification has been extensively studied in various domains, it has not received much attention in the music domain. In this paper, we investigate intent classification models for music discovery conversation, focusing on pre-trained language models. Rather than only predicting functional needs: intent classification, we also include a task for classifying musical needs: musical attribute classification. Additionally, we propose a method of concatenating previous chat history with just single-turn user queries in the input text, allowing the model to understand the overall conversation context better. Our proposed model significantly improves the F1 score for both user intent and musical attribute classification, and surpasses the zero-shot and few-shot performance of the pretrained Llama 3 model.