Authors: Konstantinos Vogiatzoglou, Costas Papadimitriou, Vasilis Bontozoglou, Konstantinos Ampountolas
Abstract: Wildland fires pose terrifying natural hazards, underscoring the urgent need to develop data-driven and physics-informed digital twins for wildfire prevention, monitoring, intervention, and response. In this direction of research, this work introduces a physics-informed neural network (PiNN) to learn the unknown parameters of an interpretable wildfire spreading model. The considered wildfire spreading model integrates fundamental physical laws articulated by key model parameters, essential for capturing the complex behavior of wildfires. The proposed machine learning approach leverages the theory of artificial neural networks with the physical constraints governing wildfire dynamics, such as the first principles of mass and energy conservation. Training of the PiNN for physics-informed parameter identification is realized using data of the temporal evolution of one- and two-dimensional (plane surface) fire fronts that have been obtained from a high-fidelity simulator of the wildfire spreading model under consideration. The parameter learning results demonstrate the remarkable predictive ability of the proposed PiNN in uncovering the unknown coefficients in both the one- and two-dimensional fire spreading scenarios. Additionally, this methodology exhibits robustness by identifying the same parameters in the presence of noisy data. The proposed framework is envisioned to be incorporated in a physics-informed digital twin for intelligent wildfire management and risk assessment.
Authors: Hao (Mark), Chen, Liam Castelli, Martin Ferianc, Hongyu Zhou, Shuanglong Liu, Wayne Luk, Hongxiang Fan
Abstract: Reliable uncertainty estimation plays a crucial role in various safety-critical applications such as medical diagnosis and autonomous driving. In recent years, Bayesian neural networks (BayesNNs) have gained substantial research and industrial interests due to their capability to make accurate predictions with reliable uncertainty estimation. However, the algorithmic complexity and the resulting hardware performance of BayesNNs hinder their adoption in real-life applications. To bridge this gap, this paper proposes an algorithm and hardware co-design framework that can generate field-programmable gate array (FPGA)-based accelerators for efficient BayesNNs. At the algorithm level, we propose novel multi-exit dropout-based BayesNNs with reduced computational and memory overheads while achieving high accuracy and quality of uncertainty estimation. At the hardware level, this paper introduces a transformation framework that can generate FPGA-based accelerators for the proposed efficient multi-exit BayesNNs. Several optimization techniques such as the mix of spatial and temporal mappings are introduced to reduce resource consumption and improve the overall hardware performance. Comprehensive experiments demonstrate that our approach can achieve higher energy efficiency compared to CPU, GPU, and other state-of-the-art hardware implementations. To support the future development of this research, we have open-sourced our code at: https://github.com/os-hxfan/MCME_FPGA_Acc.git
Authors: Juan Agustin Duque, Milad Aghajohari, Tim Cooijmans, Tianyu Zhang, Aaron Courville
Abstract: The growing presence of artificially intelligent agents in everyday decision-making, from LLM assistants to autonomous vehicles, hints at a future in which conflicts may arise from each agent optimizing individual interests. In general-sum games these conflicts are apparent, where naive Reinforcement Learning agents get stuck in Pareto-suboptimal Nash equilibria. Consequently, opponent shaping has been introduced as a method with success at finding socially beneficial equilibria in social dilemmas. In this work, we introduce Advantage Alignment, a family of algorithms derived from first principles that perform opponent shaping efficiently and intuitively. This is achieved by aligning the advantages of conflicting agents in a given game by increasing the probability of mutually-benefiting actions. We prove that existing opponent shaping methods, including LOLA and LOQA, implicitly perform Advantage Alignment. Compared to these works, Advantage Alignment mathematically simplifies the formulation of opponent shaping and seamlessly works for continuous action domains. We also demonstrate the effectiveness of our algorithm in a wide range of social dilemmas, achieving state of the art results in each case, including a social dilemma version of the Negotiation Game.
Authors: Jiarui Feng, Hao Liu, Lecheng Kong, Yixin Chen, Muhan Zhang
Abstract: In this report, we present TAGLAS, an atlas of text-attributed graph (TAG) datasets and benchmarks. TAGs are graphs with node and edge features represented in text, which have recently gained wide applicability in training graph-language or graph foundation models. In TAGLAS, we collect and integrate more than 23 TAG datasets with domains ranging from citation graphs to molecule graphs and tasks from node classification to graph question-answering. Unlike previous graph datasets and benchmarks, all datasets in TAGLAS have a unified node and edge text feature format, which allows a graph model to be simultaneously trained and evaluated on multiple datasets from various domains. Further, we provide a standardized, efficient, and simplified way to load all datasets and tasks. We also provide useful utils like text-to-embedding conversion, and graph-to-text conversion, which can facilitate different evaluation scenarios. Finally, we also provide standard and easy-to-use evaluation utils. The project is open-sourced at https://github.com/JiaruiFeng/TAGLAS and is still under construction. Please expect more datasets/features in the future.
Authors: Zhicheng Liang, Yu Yang, Xiangyu Ke, Xiaokui Xiao, Yunjun Gao
Abstract: Recent years have witnessed a growing trend toward employing deep reinforcement learning (Deep-RL) to derive heuristics for combinatorial optimization (CO) problems on graphs. Maximum Coverage Problem (MCP) and its probabilistic variant on social networks, Influence Maximization (IM), have been particularly prominent in this line of research. In this paper, we present a comprehensive benchmark study that thoroughly investigates the effectiveness and efficiency of five recent Deep-RL methods for MCP and IM. These methods were published in top data science venues, namely S2V-DQN, Geometric-QN, GCOMB, RL4IM, and LeNSE. Our findings reveal that, across various scenarios, the Lazy Greedy algorithm consistently outperforms all Deep-RL methods for MCP. In the case of IM, theoretically sound algorithms like IMM and OPIM demonstrate superior performance compared to Deep-RL methods in most scenarios. Notably, we observe an abnormal phenomenon in IM problem where Deep-RL methods slightly outperform IMM and OPIM when the influence spread nearly does not increase as the budget increases. Furthermore, our experimental results highlight common issues when applying Deep-RL methods to MCP and IM in practical settings. Finally, we discuss potential avenues for improving Deep-RL methods. Our benchmark study sheds light on potential challenges in current deep reinforcement learning research for solving combinatorial optimization problems.
Authors: Raul Astudillo, Kejun Li, Maegan Tucker, Chu Xin Cheng, Aaron D. Ames, Yisong Yue
Abstract: Preferential Bayesian optimization (PBO) is a framework for optimizing a decision-maker's latent preferences over available design choices. While preferences often involve multiple conflicting objectives, existing work in PBO assumes that preferences can be encoded by a single objective function. For example, in robotic assistive devices, technicians often attempt to maximize user comfort while simultaneously minimizing mechanical energy consumption for longer battery life. Similarly, in autonomous driving policy design, decision-makers wish to understand the trade-offs between multiple safety and performance attributes before committing to a policy. To address this gap, we propose the first framework for PBO with multiple objectives. Within this framework, we present dueling scalarized Thompson sampling (DSTS), a multi-objective generalization of the popular dueling Thompson algorithm, which may be of interest beyond the PBO setting. We evaluate DSTS across four synthetic test functions and two simulated exoskeleton personalization and driving policy design tasks, showing that it outperforms several benchmarks. Finally, we prove that DSTS is asymptotically consistent. As a direct consequence, this result provides, to our knowledge, the first convergence guarantee for dueling Thompson sampling in the PBO setting.
Authors: Milad Ramezankhani, Anirudh Deodhar, Rishi Yash Parekh, Dagnachew Birru
Abstract: Deep Operator Networks (DeepONets) and their physics-informed variants have shown significant promise in learning mappings between function spaces of partial differential equations, enhancing the generalization of traditional neural networks. However, for highly nonlinear real-world applications like aerospace composites processing, existing models often fail to capture underlying solutions accurately and are typically limited to single input functions, constraining rapid process design development. This paper introduces an advanced physics-informed DeepONet tailored for such complex systems with multiple input functions. Equipped with architectural enhancements like nonlinear decoders and effective training strategies such as curriculum learning and domain decomposition, the proposed model handles high-dimensional design spaces with significantly improved accuracy, outperforming the vanilla physics-informed DeepONet by two orders of magnitude. Its zero-shot prediction capability across a broad design space makes it a powerful tool for accelerating composites process design and optimization, with potential applications in other engineering fields characterized by strong nonlinearity.
Authors: Ti-Fen Pan, Jing-Jing Li, Bill Thompson, Anne Collins
Abstract: Extracting time-varying latent variables from computational cognitive models is a key step in model-based neural analysis, which aims to understand the neural correlates of cognitive processes. However, existing methods only allow researchers to infer latent variables that explain subjects' behavior in a relatively small class of cognitive models. For example, a broad class of relevant cognitive models with analytically intractable likelihood is currently out of reach from standard techniques, based on Maximum a Posteriori parameter estimation. Here, we present an approach that extends neural Bayes estimation to learn a direct mapping between experimental data and the targeted latent variable space using recurrent neural networks and simulated datasets. We show that our approach achieves competitive performance in inferring latent variable sequences in both tractable and intractable models. Furthermore, the approach is generalizable across different computational models and is adaptable for both continuous and discrete latent spaces. We then demonstrate its applicability in real world datasets. Our work underscores that combining recurrent neural networks and simulation-based inference to identify latent variable sequences can enable researchers to access a wider class of cognitive models for model-based neural analyses, and thus test a broader set of theories.
Authors: Wojciech Kot{\l}owski, Marek Wydmuch, Erik Schultheis, Rohit Babbar, Krzysztof Dembczy\'nski
Abstract: We consider sequential maximization of performance metrics that are general functions of a confusion matrix of a classifier (such as precision, F-measure, or G-mean). Such metrics are, in general, non-decomposable over individual instances, making their optimization very challenging. While they have been extensively studied under different frameworks in the batch setting, their analysis in the online learning regime is very limited, with only a few distinguished exceptions. In this paper, we introduce and analyze a general online algorithm that can be used in a straightforward way with a variety of complex performance metrics in binary, multi-class, and multi-label classification problems. The algorithm's update and prediction rules are appealingly simple and computationally efficient without the need to store any past data. We show the algorithm attains $\mathcal{O}(\frac{\ln n}{n})$ regret for concave and smooth metrics and verify the efficiency of the proposed algorithm in empirical studies.
Authors: Yulong Yang, Bowen Feng, Keqin Wang, Naomi Leonard, Adji Bousso Dieng, Christine Allen-Blanchette
Abstract: From pedestrians to Kuramoto oscillators, interactions between agents govern how a multitude of dynamical systems evolve in space and time. Discovering how these agents relate to each other can improve our understanding of the often complex dynamics that underlie these systems. Recent works learn to categorize relationships between agents based on observations of their physical behavior. These approaches are limited in that the relationship categories are modelled as independent and mutually exclusive, when in real world systems categories are often interacting. In this work, we introduce a level of abstraction between the physical behavior of agents and the categories that define their behavior. To do this, we learn a mapping from the agents' states to their affinities for each category in a graph neural network. We integrate the physical proximity of agents and their affinities in a nonlinear opinion dynamics model which provides a mechanism to identify mutually exclusive categories, predict an agent's evolution in time, and control an agent's behavior. We demonstrate the utility of our model for learning interpretable categories for mechanical systems, and demonstrate its efficacy on several long-horizon trajectory prediction benchmarks where we consistently out perform existing methods.
Authors: Weiqin Chen, Mark S. Squillante, Chai Wah Wu, Santiago Paternain
Abstract: We devise a control-theoretic reinforcement learning approach to support direct learning of the optimal policy. We establish theoretical properties of our approach and derive an algorithm based on a specific instance of this approach. Our empirical results demonstrate the significant benefits of our approach.
Authors: Ivana Kaji\'c, Olivia Wiles, Isabela Albuquerque, Matthias Bauer, Su Wang, Jordi Pont-Tuset, Aida Nematzadeh
Abstract: Text-to-image generative models are capable of producing high-quality images that often faithfully depict concepts described using natural language. In this work, we comprehensively evaluate a range of text-to-image models on numerical reasoning tasks of varying difficulty, and show that even the most advanced models have only rudimentary numerical skills. Specifically, their ability to correctly generate an exact number of objects in an image is limited to small numbers, it is highly dependent on the context the number term appears in, and it deteriorates quickly with each successive number. We also demonstrate that models have poor understanding of linguistic quantifiers (such as "a few" or "as many as"), the concept of zero, and struggle with more advanced concepts such as partial quantities and fractional representations. We bundle prompts, generated images and human annotations into GeckoNum, a novel benchmark for evaluation of numerical reasoning.
Authors: Alexandre Jacquillat, Michael Lingzhi Li
Abstract: We define an online learning and optimization problem with irreversible decisions contributing toward a coverage target. At each period, a decision-maker selects facilities to open, receives information on the success of each one, and updates a machine learning model to guide future decisions. The goal is to minimize costs across a finite horizon under a chance constraint reflecting the coverage target. We derive an optimal algorithm and a tight lower bound in an asymptotic regime characterized by a large target number of facilities $m\to\infty$ but a finite horizon $T\in\mathbb{Z}_+$. We find that the regret grows sub-linearly at a rate $\Theta\left(m^{\frac{1}{2}\cdot\frac{1}{1-2^{-T}}}\right)$, thus converging exponentially fast to $\Theta(\sqrt{m})$. We establish the robustness of this result to the learning environment; we also extend it to a more complicated facility location setting in a bipartite facility-customer graph with a target on customer coverage. Throughout, constructive proofs identify a policy featuring limited exploration initially for learning purposes, and fast exploitation later on for optimization purposes once uncertainty gets mitigated. These findings underscore the benefits of limited online learning and optimization, in that even a few rounds can provide significant benefits as compared to a no-learning baseline.
Authors: Riddhiman Bhattacharya, Thanh Nguyen, Will Wei Sun, Mohit Tawarmalani
Abstract: We explore an active learning approach for dynamic fair resource allocation problems. Unlike previous work that assumes full feedback from all agents on their allocations, we consider feedback from a select subset of agents at each epoch of the online resource allocation process. Despite this restriction, our proposed algorithms provide regret bounds that are sub-linear in number of time-periods for various measures that include fairness metrics commonly used in resource allocation problems and stability considerations in matching mechanisms. The key insight of our algorithms lies in adaptively identifying the most informative feedback using dueling upper and lower confidence bounds. With this strategy, we show that efficient decision-making does not require extensive feedback and produces efficient outcomes for a variety of problem classes.
Authors: Max Wasserman, Gonzalo Mateos
Abstract: Graphs serve as generic tools to encode the underlying relational structure of data. Often this graph is not given, and so the task of inferring it from nodal observations becomes important. Traditional approaches formulate a convex inverse problem with a smoothness promoting objective and rely on iterative methods to obtain a solution. In supervised settings where graph labels are available, one can unroll and truncate these iterations into a deep network that is trained end-to-end. Such a network is parameter efficient and inherits inductive bias from the optimization formulation, an appealing aspect for data constrained settings in, e.g., medicine, finance, and the natural sciences. But typically such settings care equally about uncertainty over edge predictions, not just point estimates. Here we introduce novel iterations with independently interpretable parameters, i.e., parameters whose values - independent of other parameters' settings - proportionally influence characteristics of the estimated graph, such as edge sparsity. After unrolling these iterations, prior knowledge over such graph characteristics shape prior distributions over these independently interpretable network parameters to yield a Bayesian neural network (BNN) capable of graph structure learning (GSL) from smooth signal observations. Fast execution and parameter efficiency allow for high-fidelity posterior approximation via Markov Chain Monte Carlo (MCMC) and thus uncertainty quantification on edge predictions. Synthetic and real data experiments corroborate this model's ability to provide well-calibrated estimates of uncertainty, in test cases that include unveiling economic sector modular structure from S$\&$P$500$ data and recovering pairwise digit similarities from MNIST images. Overall, this framework enables GSL in modest-scale applications where uncertainty on the data structure is paramount.
Authors: Jiali Cheng, Hadi Amiri
Abstract: Recent advancements in Machine Unlearning (MU) have introduced solutions to selectively remove certain training samples, such as those with outdated or sensitive information, from trained models. Despite these advancements, evaluation of MU methods have been inconsistent, employing different trained models and architectures, and sample removal strategies, which hampers accurate comparison. In addition, prior MU approaches have mainly focused on singular tasks or modalities, which is not comprehensive. To address these limitations, we develop MU-Bench, the first comprehensive benchmark for MU that (i) unifies the sets of deleted samples and trained models, and (ii) provides broad coverage of tasks and data modalities, including previously unexplored domains such as speech and video classification. Our evaluation show that RandLabel and SalUn are the most effective general MU approaches on MU-Bench, and BadT and SCRUB are capable of achieving random performance on the deletion set. We analyze several under-investigated aspects of unlearning, including scalability, the impacts of parameter-efficient fine-tuning and curriculum learning, and susceptibility to dataset biases. MU-Bench provides an easy-to-use package that includes dataset splits, models, and implementations, together with a leader board to enable unified and scalable MU research.
Authors: Salva R\"uhling Cachay, Brian Henn, Oliver Watt-Meyer, Christopher S. Bretherton, Rose Yu
Abstract: Data-driven deep learning models are on the verge of transforming global weather forecasting. It is an open question if this success can extend to climate modeling, where long inference rollouts and data complexity pose significant challenges. Here, we present the first conditional generative model able to produce global climate ensemble simulations that are accurate and physically consistent. Our model runs at 6-hourly time steps and is shown to be stable for 10-year-long simulations. Our approach beats relevant baselines and nearly reaches a gold standard for successful climate model emulation. We discuss the key design choices behind our dynamics-informed diffusion model-based approach which enables this significant step towards efficient, data-driven climate simulations that can help us better understand the Earth and adapt to a changing climate.
Authors: Jingyi Liu, Yanjie Li, Lina Yu, Min Wu, Weijun Li, Wenqiang Li, Meilan Hao, Yusong Deng, Shu Wei
Abstract: Noise ubiquitously exists in signals due to numerous factors including physical, electronic, and environmental effects. Traditional methods of symbolic regression, such as genetic programming or deep learning models, aim to find the most fitting expressions for these signals. However, these methods often overlook the noise present in real-world data, leading to reduced fitting accuracy. To tackle this issue, we propose \textit{\textbf{D}eep Symbolic Regression against \textbf{N}oise via \textbf{C}ontrastive \textbf{L}earning (DN-CL)}. DN-CL employs two parameter-sharing encoders to embed data points from various data transformations into feature shields against noise. This model treats noisy data and clean data as different views of the ground-truth mathematical expressions. Distances between these features are minimized, utilizing contrastive learning to distinguish between 'positive' noise-corrected pairs and 'negative' contrasting pairs. Our experiments indicate that DN-CL demonstrates superior performance in handling both noisy and clean data, presenting a promising method of symbolic regression.
Authors: Bo Jiang, Sheng Ge, Ziyan Zhang, Beibei Wang, Jin Tang, Bin Luo
Abstract: Graph Convolutional Networks (GCNs) have been widely studied. The core of GCNs is the definition of convolution operators on graphs. However, existing Graph Convolution (GC) operators are mainly defined on adjacency matrix and node features and generally focus on obtaining effective node embeddings which cannot be utilized to address the graphs with (high-dimensional) edge features. To address this problem, by leveraging tensor contraction representation and tensor product graph diffusion theories, this paper analogously defines an effective convolution operator on graphs with edge features which is named as Tensor Product Graph Convolution (TPGC). The proposed TPGC aims to obtain effective edge embeddings. It provides a complementary model to traditional graph convolutions (GCs) to address the more general graph data analysis with both node and edge features. Experimental results on several graph learning tasks demonstrate the effectiveness of the proposed TPGC.
Authors: Mengdan Zhu, Raasikh Kanjiani, Jiahui Lu, Andrew Choi, Qirui Ye, Liang Zhao
Abstract: Deep generative models like VAEs and diffusion models have advanced various generation tasks by leveraging latent variables to learn data distributions and generate high-quality samples. Despite the field of explainable AI making strides in interpreting machine learning models, understanding latent variables in generative models remains challenging. This paper introduces LatentExplainer, a framework for automatically generating semantically meaningful explanations of latent variables in deep generative models. LatentExplainer tackles three main challenges: inferring the meaning of latent variables, aligning explanations with inductive biases, and handling varying degrees of explainability. By perturbing latent variables and interpreting changes in generated data, the framework provides a systematic approach to understanding and controlling the data generation process, enhancing the transparency and interpretability of deep generative models. We evaluate our proposed method on several real-world and synthetic datasets, and the results demonstrate superior performance in generating high-quality explanations of latent variables.
Authors: Sisi Shao, Pedro Henrique Ribeiro, Christina Ramirez, Jason H. Moore
Abstract: Feature selection in Knowledge Graphs (KGs) are increasingly utilized in diverse domains, including biomedical research, Natural Language Processing (NLP), and personalized recommendation systems. This paper delves into the methodologies for feature selection within KGs, emphasizing their roles in enhancing machine learning (ML) model efficacy, hypothesis generation, and interpretability. Through this comprehensive review, we aim to catalyze further innovation in feature selection for KGs, paving the way for more insightful, efficient, and interpretable analytical models across various domains. Our exploration reveals the critical importance of scalability, accuracy, and interpretability in feature selection techniques, advocating for the integration of domain knowledge to refine the selection process. We highlight the burgeoning potential of multi-objective optimization and interdisciplinary collaboration in advancing KG feature selection, underscoring the transformative impact of such methodologies on precision medicine, among other fields. The paper concludes by charting future directions, including the development of scalable, dynamic feature selection algorithms and the integration of explainable AI principles to foster transparency and trust in KG-driven models.
Authors: Kyle Wong, Alfonso Amayuelas, Liangming Pan, William Yang Wang
Abstract: Large language models (LLMs) have shown remarkable performance on code generation tasks. A recent application of LLMs for code generation is iterative code repair, where a model fixes an incorrect program by rationalizing about errors and generating a new program. However, code repair is primarily studied on high-resource languages like Python, and the framework's efficacy is under-explored on low-resource languages. To apply code repair for low-resource languages, we propose Distilling Low-Resource Repairs (DistiLRR), an approach that transfers the reasoning and code generation ability from a teacher model to a student model. Our results show that DistiLRR consistently outperforms baselines on low-resource languages, but has similar performance on high-resource languages. To investigate this behavior, we perform a further analysis and find that the correlation between rationale quality and code correctness is weaker than previously perceived. We hypothesize this weakness is magnified in low-resource settings where base models lack deep knowledge of a programming language, leading to wavering benefits of code repair between high-resource and low-resource languages.
Authors: Deokjae Lee, Hyun Oh Song, Kyunghyun Cho
Abstract: Active learning is increasingly adopted for expensive multi-objective combinatorial optimization problems, but it involves a challenging subset selection problem, optimizing the batch acquisition score that quantifies the goodness of a batch for evaluation. Due to the excessively large search space of the subset selection problem, prior methods optimize the batch acquisition on the latent space, which has discrepancies with the actual space, or optimize individual acquisition scores without considering the dependencies among candidates in a batch instead of directly optimizing the batch acquisition. To manage the vast search space, a simple and effective approach is the greedy method, which decomposes the problem into smaller subproblems, yet it has difficulty in parallelization since each subproblem depends on the outcome from the previous ones. To this end, we introduce a novel greedy-style subset selection algorithm that optimizes batch acquisition directly on the combinatorial space by sequential greedy sampling from the greedy policy, specifically trained to address all greedy subproblems concurrently. Notably, our experiments on the red fluorescent proteins design task show that our proposed method achieves the baseline performance in 1.69x fewer queries, demonstrating its efficiency.
Authors: Chongzhi Zhang, Zhiping Peng, Junhao Zheng, Linghao Wang, Ruifeng Shi, Qianli Ma
Abstract: Complex Logical Query Answering (CLQA) over incomplete knowledge graphs is a challenging task. Recently, Query Embedding (QE) methods are proposed to solve CLQA by performing multi-hop logical reasoning. However, most of them only consider historical query context information while ignoring future information, which leads to their failure to capture the complex dependencies behind the elements of a query. In recent years, the transformer architecture has shown a strong ability to model long-range dependencies between words. The bidirectional attention mechanism proposed by the transformer can solve the limitation of these QE methods regarding query context. Still, as a sequence model, it is difficult for the transformer to model complex logical queries with branch structure computation graphs directly. To this end, we propose a neural one-point embedding method called Pathformer based on the tree-like computation graph, i.e., query computation tree. Specifically, Pathformer decomposes the query computation tree into path query sequences by branches and then uses the transformer encoder to recursively encode these path query sequences to obtain the final query embedding. This allows Pathformer to fully utilize future context information to explicitly model the complex interactions between various parts of the path query. Experimental results show that Pathformer outperforms existing competitive neural QE methods, and we found that Pathformer has the potential to be applied to non-one-point embedding space.
Authors: Tianyu Fu, Haofeng Huang, Xuefei Ning, Genghan Zhang, Boju Chen, Tianqi Wu, Hongyi Wang, Zixiao Huang, Shiyao Li, Shengen Yan, Guohao Dai, Huazhong Yang, Yu Wang
Abstract: Sparse attention can effectively mitigate the significant memory and throughput demands of Large Language Models (LLMs) in long contexts. Existing methods typically employ a uniform sparse attention mask, applying the same sparse pattern across different attention heads and input lengths. However, this uniform approach fails to capture the diverse attention patterns inherent in LLMs, ignoring their distinct accuracy-latency trade-offs. To address this challenge, we propose the Mixture of Attention (MoA), which automatically tailors distinct sparse attention configurations to different heads and layers. MoA constructs and navigates a search space of various attention patterns and their scaling rules relative to input sequence lengths. It profiles the model, evaluates potential configurations, and pinpoints the optimal sparse attention compression plan. MoA adapts to varying input sizes, revealing that some attention heads expand their focus to accommodate longer sequences, while other heads consistently concentrate on fixed-length local contexts. Experiments show that MoA increases the effective context length by $3.9\times$ with the same average attention span, boosting retrieval accuracy by $1.5-7.1\times$ over the uniform-attention baseline across Vicuna-7B, Vicuna-13B, and Llama3-8B models. Moreover, MoA narrows the capability gaps between sparse and dense models, reducing the maximum relative performance drop from $9\%-36\%$ to within $5\%$ across two long-context understanding benchmarks. MoA achieves a $1.2-1.4\times$ GPU memory reduction and boosts decode throughput by $5.5-6.7 \times$ for 7B and 13B dense models on a single GPU, with minimal impact on performance.
Authors: Xiaojing Chen, Zhenyuan Li, Wei Ni, Xin Wang, Shunqing Zhang, Yanzan Sun, Shugong Xu, Qingqi Pei
Abstract: Federated learning (FL) is a viable technique to train a shared machine learning model without sharing data. Hierarchical FL (HFL) system has yet to be studied regrading its multiple levels of energy, computation, communication, and client scheduling, especially when it comes to clients relying on energy harvesting to power their operations. This paper presents a new two-phase deep deterministic policy gradient (DDPG) framework, referred to as ``TP-DDPG'', to balance online the learning delay and model accuracy of an FL process in an energy harvesting-powered HFL system. The key idea is that we divide optimization decisions into two groups, and employ DDPG to learn one group in the first phase, while interpreting the other group as part of the environment to provide rewards for training the DDPG in the second phase. Specifically, the DDPG learns the selection of participating clients, and their CPU configurations and the transmission powers. A new straggler-aware client association and bandwidth allocation (SCABA) algorithm efficiently optimizes the other decisions and evaluates the reward for the DDPG. Experiments demonstrate that with substantially reduced number of learnable parameters, the TP-DDPG can quickly converge to effective polices that can shorten the training time of HFL by 39.4% compared to its benchmarks, when the required test accuracy of HFL is 0.9.
Authors: Wei Zhuo, Guang Tan
Abstract: We consider the graph similarity computation (GSC) task based on graph edit distance (GED) estimation. State-of-the-art methods treat GSC as a learning-based prediction task using Graph Neural Networks (GNNs). To capture fine-grained interactions between pair-wise graphs, these methods mostly contain a node-level matching module in the end-to-end learning pipeline, which causes high computational costs in both the training and inference stages. We show that the expensive node-to-node matching module is not necessary for GSC, and high-quality learning can be attained with a simple yet powerful regularization technique, which we call the Alignment Regularization (AReg). In the training stage, the AReg term imposes a node-graph correspondence constraint on the GNN encoder. In the inference stage, the graph-level representations learned by the GNN encoder are directly used to compute the similarity score without using AReg again to speed up inference. We further propose a multi-scale GED discriminator to enhance the expressive ability of the learned representations. Extensive experiments on real-world datasets demonstrate the effectiveness, efficiency and transferability of our approach.
Authors: Erion Morina, Martin Holler
Abstract: This work focuses on the analysis of fully connected feed forward ReLU neural networks as they approximate a given, smooth function. In contrast to conventionally studied universal approximation properties under increasing architectures, e.g., in terms of width or depth of the networks, we are concerned with the asymptotic growth of the parameters of approximating networks. Such results are of interest, e.g., for error analysis or consistency results for neural network training. The main result of our work is that, for a ReLU architecture with state of the art approximation error, the realizing parameters grow at most polynomially. The obtained rate with respect to a normalized network size is compared to existing results and is shown to be superior in most cases, in particular for high dimensional input.
Authors: Kris De Asis, Richard S. Sutton
Abstract: Many reinforcement learning algorithms are built on an assumption that an agent interacts with an environment over fixed-duration, discrete time steps. However, physical systems are continuous in time, requiring a choice of time-discretization granularity when digitally controlling them. Furthermore, such systems do not wait for decisions to be made before advancing the environment state, necessitating the study of how the choice of discretization may affect a reinforcement learning algorithm. In this work, we consider the relationship between the definitions of the continuous-time and discrete-time returns. Specifically, we acknowledge an idiosyncrasy with naively applying a discrete-time algorithm to a discretized continuous-time environment, and note how a simple modification can better align the return definitions. This observation is of practical consideration when dealing with environments where time-discretization granularity is a choice, or situations where such granularity is inherently stochastic.
Authors: Zixi Zhang, Cheng Zhang, Xitong Gao, Robert D. Mullins, George A. Constantinides, Yiren Zhao
Abstract: Low-rank Adaption (LoRA) has been the de-facto parameter-efficient fine-tuning technique for large language models. We present HeteroLoRA, a light-weight search algorithm that leverages zero-cost proxies to allocate the limited LoRA trainable parameters across the model for better fine-tuned performance. In addition to the allocation for the standard LoRA-adapted models, we also demonstrate the efficacy of HeteroLoRA by performing the allocation in a more challenging search space that includes LoRA modules and LoRA-adapted shortcut connections. Experiments show that HeteroLoRA enables improvements in model performance given the same parameter budge. For example, on MRPC, we see an improvement of 1.6% in accuracy with similar training parameter budget. We will open-source our algorithm once the paper is accepted.
Authors: Yuang Chen, Cheng Zhang, Xitong Gao, Robert D. Mullins, George A. Constantinides, Yiren Zhao
Abstract: Grouped-query attention (GQA) has been widely adopted in LLMs to mitigate the complexity of multi-head attention (MHA). To transform an MHA to a GQA, neighbour queries in MHA are evenly split into groups where each group shares the value and key layers. In this work, we propose AsymGQA, an activation-informed approach to asymmetrically grouping an MHA to a GQA for better model performance. Our AsymGQA outperforms the GQA within the same model size budget. For example, AsymGQA LLaMA-2-7B has an accuracy increase of 7.5% on MMLU compared to neighbour grouping. Our approach addresses the GQA's trade-off problem between model performance and hardware efficiency.
Authors: Xiaohong Ji, Wang Zhen, Zhifeng Gao, Hang Zheng, Linfeng Zhang, Guolin Ke, Weinan E
Abstract: In recent years, pretraining models have made significant advancements in the fields of natural language processing (NLP), computer vision (CV), and life sciences. The significant advancements in NLP and CV are predominantly driven by the expansion of model parameters and data size, a phenomenon now recognized as the scaling laws. However, research exploring scaling law in molecular pretraining models remains unexplored. In this work, we present Uni-Mol2 , an innovative molecular pretraining model that leverages a two-track transformer to effectively integrate features at the atomic level, graph level, and geometry structure level. Along with this, we systematically investigate the scaling law within molecular pretraining models, characterizing the power-law correlations between validation loss and model size, dataset size, and computational resources. Consequently, we successfully scale Uni-Mol2 to 1.1 billion parameters through pretraining on 800 million conformations, making it the largest molecular pretraining model to date. Extensive experiments show consistent improvement in the downstream tasks as the model size grows. The Uni-Mol2 with 1.1B parameters also outperforms existing methods, achieving an average 27% improvement on the QM9 and 14% on COMPAS-1D dataset.
Authors: Arsentii Kuzmin, Alexander Aduenko, Vadim Strijov
Abstract: In this paper, we develop a decision support system for the hierarchical text classification. We consider text collections with a fixed hierarchical structure of topics given by experts in the form of a tree. The system sorts the topics by relevance to a given document. The experts choose one of the most relevant topics to finish the classification. We propose a weighted hierarchical similarity function to calculate topic relevance. The function calculates the similarity of a document and a tree branch. The weights in this function determine word importance. We use the entropy of words to estimate the weights. The proposed hierarchical similarity function formulates a joint hierarchical thematic classification probability model of the document topics, parameters, and hyperparameters. The variational Bayesian inference gives a closed-form EM algorithm. The EM algorithm estimates the parameters and calculates the probability of a topic for a given document. Compared to hierarchical multiclass SVM, hierarchical PLSA with adaptive regularization, and hierarchical naive Bayes, the weighted hierarchical similarity function has better improvement in ranking accuracy in an abstract collection of a major conference EURO and a website collection of industrial companies.
Authors: Thomas Hehn, Markus Peschl, Tribhuvanesh Orekondy, Arash Behboodi, Johann Brehmer
Abstract: Modelling the propagation of electromagnetic signals is critical for designing modern communication systems. While there are precise simulators based on ray tracing, they do not lend themselves to solving inverse problems or the integration in an automated design loop. We propose to address these challenges through differentiable neural surrogates that exploit the geometric aspects of the problem. We first introduce the Wireless Geometric Algebra Transformer (Wi-GATr), a generic backbone architecture for simulating wireless propagation in a 3D environment. It uses versatile representations based on geometric algebra and is equivariant with respect to E(3), the symmetry group of the underlying physics. Second, we study two algorithmic approaches to signal prediction and inverse problems based on differentiable predictive modelling and diffusion models. We show how these let us predict received power, localize receivers, and reconstruct the 3D environment from the received signal. Finally, we introduce two large, geometry-focused datasets of wireless signal propagation in indoor scenes. In experiments, we show that our geometry-forward approach achieves higher-fidelity predictions with less data than various baselines.
Authors: Matthias Weissenbacher, Rishabh Agarwal, Yoshinobu Kawahara
Abstract: An open challenge in reinforcement learning (RL) is the effective deployment of a trained policy to new or slightly different situations as well as semantically-similar environments. We introduce Symmetry-Invariant Transformer (SiT), a scalable vision transformer (ViT) that leverages both local and global data patterns in a self-supervised manner to improve generalisation. Central to our approach is Graph Symmetric Attention, which refines the traditional self-attention mechanism to preserve graph symmetries, resulting in invariant and equivariant latent representations. We showcase SiT's superior generalization over ViTs on MiniGrid and Procgen RL benchmarks, and its sample efficiency on Atari 100k and CIFAR10.
Authors: Jack R. P. Hanslope, Laurence Aitchison
Abstract: In climate science, we often want to compare across different datasets. Difficulties can arise in doing this due to inevitable mismatches that arise between observational and reanalysis data, or even between different reanalyses. This misalignment can raise problems for any work that seeks to make inferences about one dataset from another. We considered tropical cyclone location as an example task with one dataset providing atmospheric conditions (ERA5) and another providing storm tracks (IBTrACS). We found that while the examples often aligned well, there were a considerable proportion (around 25%) which were not well aligned. We trained a neural network to map from the wind field to the storm location; in this setting misalignment in the datasets appears as "label noise" (i.e. the labelled storm location does not correspond to the underlying wind field). We found that this neural network trained only on the often noisy labels from IBTrACS had a denoising effect, and performed better than the IBTrACS labels themselves, as measured by human preferences. Remarkably, this even held true for training points, on which we might have expected the network to overfit to the IBTrACS predictions.
Authors: Francisco de Arriba-P\'erez, Silvia Garc\'ia-M\'endez, F\'atima Leal, Benedita Malheiro, J. C. Burguillo
Abstract: Spam reviews are a pervasive problem on online platforms due to its significant impact on reputation. However, research into spam detection in data streams is scarce. Another concern lies in their need for transparency. Consequently, this paper addresses those problems by proposing an online solution for identifying and explaining spam reviews, incorporating data drift adaptation. It integrates (i) incremental profiling, (ii) data drift detection & adaptation, and (iii) identification of spam reviews employing Machine Learning. The explainable mechanism displays a visual and textual prediction explanation in a dashboard. The best results obtained reached up to 87 % spam F-measure.
Authors: Andrei Lupu, Chris Lu, Jarek Liesen, Robert Tjarko Lange, Jakob Foerster
Abstract: Dataset distillation aims to condense large datasets into a small number of synthetic examples that can be used as drop-in replacements when training new models. It has applications to interpretability, neural architecture search, privacy, and continual learning. Despite strong successes in supervised domains, such methods have not yet been extended to reinforcement learning, where the lack of a fixed dataset renders most distillation methods unusable. Filling the gap, we formalize behaviour distillation, a setting that aims to discover and then condense the information required for training an expert policy into a synthetic dataset of state-action pairs, without access to expert data. We then introduce Hallucinating Datasets with Evolution Strategies (HaDES), a method for behaviour distillation that can discover datasets of just four state-action pairs which, under supervised learning, train agents to competitive performance levels in continuous control tasks. We show that these datasets generalize out of distribution to training policies with a wide range of architectures and hyperparameters. We also demonstrate application to a downstream task, namely training multi-task agents in a zero-shot fashion. Beyond behaviour distillation, HaDES provides significant improvements in neuroevolution for RL over previous approaches and achieves SoTA results on one standard supervised dataset distillation task. Finally, we show that visualizing the synthetic datasets can provide human-interpretable task insights.
Authors: Qi Zhang, Mingfei Lu, Shujian Yu, Jingmin Xin, Badong Chen
Abstract: We introduce an innovative and mathematically rigorous definition for computing common information from multi-view data, drawing inspiration from G\'acs-K\"orner common information in information theory. Leveraging this definition, we develop a novel supervised multi-view learning framework to capture both common and unique information. By explicitly minimizing a total correlation term, the extracted common information and the unique information from each view are forced to be independent of each other, which, in turn, theoretically guarantees the effectiveness of our framework. To estimate information-theoretic quantities, our framework employs matrix-based R{\'e}nyi's $\alpha$-order entropy functional, which forgoes the need for variational approximation and distributional estimation in high-dimensional space. Theoretical proof is provided that our framework can faithfully discover both common and unique information from multi-view data. Experiments on synthetic and seven benchmark real-world datasets demonstrate the superior performance of our proposed framework over state-of-the-art approaches.
Authors: Adnan Ali, Jinlong Li, Huanhuan Chen, Ali Kashif Bashir
Abstract: Graph contrastive learning (GCL) aims to contrast positive-negative counterparts to learn the node embeddings, whereas graph data augmentation methods are employed to generate these positive-negative samples. The variation, quantity, and quality of negative samples compared to positive samples play crucial roles in learning meaningful embeddings for node classification downstream tasks. Less variation, excessive quantity, and low-quality negative samples cause the model to be overfitted for particular nodes, resulting in less robust models. To solve the overfitting problem in the GCL paradigm, this study proposes a novel Cumulative Sample Selection (CSS) algorithm by comprehensively considering negative samples' quality, variations, and quantity. Initially, three negative sample pools are constructed: easy, medium, and hard negative samples, which contain 25%, 50%, and 25% of the total available negative samples, respectively. Then, 10% negative samples are selected from each of these three negative sample pools for training the model. After that, a decision agent module evaluates model training results and decides whether to explore more negative samples from three negative sample pools by increasing the ratio or keep exploiting the current sampling ratio. The proposed algorithm is integrated into a proposed graph contrastive learning framework named NegAmplify. NegAmplify is compared with the SOTA methods on nine graph node classification datasets, with seven achieving better node classification accuracy with up to 2.86% improvement.
Authors: Lin Fan, Xun Gong, Cenyang Zheng, Yafei Ou
Abstract: The intersection of medical Visual Question Answering (Med-VQA) is a challenging research topic with advantages including patient engagement and clinical expert involvement for second opinions. However, existing Med-VQA methods based on joint embedding fail to explain whether their provided results are based on correct reasoning or coincidental answers, which undermines the credibility of VQA answers. In this paper, we investigate the construction of a more cohesive and stable Med-VQA structure. Motivated by causal effect, we propose a novel Triangular Reasoning VQA (Tri-VQA) framework, which constructs reverse causal questions from the perspective of "Why this answer?" to elucidate the source of the answer and stimulate more reasonable forward reasoning processes. We evaluate our method on the Endoscopic Ultrasound (EUS) multi-attribute annotated dataset from five centers, and test it on medical VQA datasets. Experimental results demonstrate the superiority of our approach over existing methods. Our codes and pre-trained models are available at https://anonymous.4open.science/r/Tri_VQA.
Authors: Valentino Maiorca, Luca Moschella, Marco Fumero, Francesco Locatello, Emanuele Rodol\`a
Abstract: The emergence of similar representations between independently trained neural models has sparked significant interest in the representation learning community, leading to the development of various methods to obtain communication between latent spaces. "Latent space communication" can be achieved in two ways: i) by independently mapping the original spaces to a shared or relative one; ii) by directly estimating a transformation from a source latent space to a target one. In this work, we combine the two into a novel method to obtain latent space translation through the relative space. By formalizing the invertibility of angle-preserving relative representations and assuming the scale invariance of decoder modules in neural models, we can effectively use the relative space as an intermediary, independently projecting onto and from other semantically similar spaces. Extensive experiments over various architectures and datasets validate our scale invariance assumption and demonstrate the high accuracy of our method in latent space translation. We also apply our method to zero-shot stitching between arbitrary pre-trained text and image encoders and their classifiers, even across modalities. Our method has significant potential for facilitating the reuse of models in a practical manner via compositionality.
Authors: Matthieu Blanke, Ronan Fablet, Marc Lelarge
Abstract: Data assimilation is a central problem in many geophysical applications, such as weather forecasting. It aims to estimate the state of a potentially large system, such as the atmosphere, from sparse observations, supplemented by prior physical knowledge. The size of the systems involved and the complexity of the underlying physical equations make it a challenging task from a computational point of view. Neural networks represent a promising method of emulating the physics at low cost, and therefore have the potential to considerably improve and accelerate data assimilation. In this work, we introduce a deep learning approach where the physical system is modeled as a sequence of coarse-to-fine Gaussian prior distributions parametrized by a neural network. This allows us to define an assimilation operator, which is trained in an end-to-end fashion to minimize the reconstruction error on a dataset with different observation processes. We illustrate our approach on chaotic dynamical physical systems with sparse observations, and compare it to traditional variational data assimilation methods.
Authors: Darko Drakulic, Sofia Michel, Jean-Marc Andreoli
Abstract: Machine Learning-based heuristics have recently shown impressive performance in solving a variety of hard combinatorial optimization problems (COPs). However they generally rely on a separate neural model, specialized and trained for each single problem. Any variation of a problem requires adjustment of its model and re-training from scratch. In this paper, we propose GOAL (for Generalist combinatorial Optimization Agent Learning), a generalist model capable of efficiently solving multiple COPs and which can be fine-tuned to solve new COPs. GOAL consists of a single backbone plus light-weight problem-specific adapters, mostly for input and output processing. The backbone is based on a new form of mixed-attention blocks which allows to handle problems defined on graphs with arbitrary combinations of node, edge and instance-level features. Additionally, problems which involve heterogeneous nodes or edges, such as in multi-partite graphs, are handled through a novel multi-type transformer architecture, where the attention blocks are duplicated to attend only the relevant combination of types while relying on the same shared parameters. We train GOAL on a set of routing, scheduling and classic graph problems and show that it is only slightly inferior to the specialized baselines while being the first multi-task model that solves a variety of COPs. Finally, we showcase the strong transfer learning capacity of GOAL by fine-tuning or learning the adapters for new problems, with only few shots and little data.
Authors: Simon Dietz, Thomas Altstidl, Dario Zanca, Bj\"orn Eskofier, An Nguyen
Abstract: Mixed-type time series (MTTS) is a bimodal data type that is common in many domains, such as healthcare, finance, environmental monitoring, and social media. It consists of regularly sampled continuous time series and irregularly sampled categorical event sequences. The integration of both modalities through multimodal fusion is a promising approach for processing MTTS. However, the question of how to effectively fuse both modalities remains open. In this paper, we present a comprehensive evaluation of several deep multimodal fusion approaches for MTTS forecasting. Our comparison includes three fusion types (early, intermediate, and late) and five fusion methods (concatenation, weighted mean, weighted mean with correlation, gating, and feature sharing). We evaluate these fusion approaches on three distinct datasets, one of which was generated using a novel framework. This framework allows for the control of key data properties, such as the strength and direction of intermodal interactions, modality imbalance, and the degree of randomness in each modality, providing a more controlled environment for testing fusion approaches. Our findings show that the performance of different fusion approaches can be substantially influenced by the direction and strength of intermodal interactions. The study reveals that early and intermediate fusion approaches excel at capturing fine-grained and coarse-grained cross-modal features, respectively. These findings underscore the crucial role of intermodal interactions in determining the most effective fusion strategy for MTTS forecasting.
Authors: Gianluca Drappo, Alberto Maria Metelli, Marcello Restelli
Abstract: Hierarchical Reinforcement Learning (HRL) approaches have shown successful results in solving a large variety of complex, structured, long-horizon problems. Nevertheless, a full theoretical understanding of this empirical evidence is currently missing. In the context of the \emph{option} framework, prior research has devised efficient algorithms for scenarios where options are fixed, and the high-level policy selecting among options only has to be learned. However, the fully realistic scenario in which both the high-level and the low-level policies are learned is surprisingly disregarded from a theoretical perspective. This work makes a step towards the understanding of this latter scenario. Focusing on the finite-horizon problem, we present a meta-algorithm alternating between regret minimization algorithms instanced at different (high and low) temporal abstractions. At the higher level, we treat the problem as a Semi-Markov Decision Process (SMDP), with fixed low-level policies, while at a lower level, inner option policies are learned with a fixed high-level policy. The bounds derived are compared with the lower bound for non-hierarchical finite-horizon problems, allowing to characterize when a hierarchical approach is provably preferable, even without pre-trained options.
Authors: Sunwoo Lee, Tuo Zhang, Saurav Prakash, Yue Niu, Salman Avestimehr
Abstract: In Federated Learning (FL), clients may have weak devices that cannot train the full model or even hold it in their memory space. To implement large-scale FL applications, thus, it is crucial to develop a distributed learning method that enables the participation of such weak clients. We propose EmbracingFL, a general FL framework that allows all available clients to join the distributed training regardless of their system resource capacity. The framework is built upon a novel form of partial model training method in which each client trains as many consecutive output-side layers as its system resources allow. Our study demonstrates that EmbracingFL encourages each layer to have similar data representations across clients, improving FL efficiency. The proposed partial model training method guarantees convergence to a neighbor of stationary points for non-convex and smooth problems. We evaluate the efficacy of EmbracingFL under a variety of settings with a mixed number of strong, moderate (~40% memory), and weak (~15% memory) clients, datasets (CIFAR-10, FEMNIST, and IMDB), and models (ResNet20, CNN, and LSTM). Our empirical study shows that EmbracingFL consistently achieves high accuracy as like all clients are strong, outperforming the state-of-the-art width reduction methods (i.e. HeteroFL and FjORD).
Authors: Philipp Becker, Niklas Freymuth, Gerhard Neumann
Abstract: Probabilistic State Space Models (SSMs) are essential for Reinforcement Learning (RL) from high-dimensional, partial information as they provide concise representations for control. Yet, they lack the computational efficiency of their recent deterministic counterparts such as S4 or Mamba. We propose KalMamba, an efficient architecture to learn representations for RL that combines the strengths of probabilistic SSMs with the scalability of deterministic SSMs. KalMamba leverages Mamba to learn the dynamics parameters of a linear Gaussian SSM in a latent space. Inference in this latent space amounts to standard Kalman filtering and smoothing. We realize these operations using parallel associative scanning, similar to Mamba, to obtain a principled, highly efficient, and scalable probabilistic SSM. Our experiments show that KalMamba competes with state-of-the-art SSM approaches in RL while significantly improving computational efficiency, especially on longer interaction sequences.
Authors: Zhengxin Yang, Wanling Gao, Luzhou Peng, Yunyou Huang, Fei Tang, Jianfeng Zhan
Abstract: Designing and optimizing neural network architectures typically requires extensive expertise, starting with handcrafted designs and then manual or automated refinement. This dependency presents a significant barrier to rapid innovation. Recognizing the complexity of automatically generating neural network architecture from scratch, we introduce Younger, a pioneering dataset to advance this ambitious goal. Derived from over 174K real-world models across more than 30 tasks from various public model hubs, Younger includes 7,629 unique architectures, and each is represented as a directed acyclic graph with detailed operator-level information. The dataset facilitates two primary design paradigms: global, for creating complete architectures from scratch, and local, for detailed architecture component refinement. By establishing these capabilities, Younger contributes to a new frontier, Artificial Intelligence-Generated Neural Network Architecture (AIGNNA). Our experiments explore the potential and effectiveness of Younger for automated architecture generation and, as a secondary benefit, demonstrate that Younger can serve as a benchmark dataset, advancing the development of graph neural networks. We release the dataset and code publicly to lower the entry barriers and encourage further research in this challenging area.
Authors: Alona Levy-Jurgenson, Zohar Yakhini
Abstract: Generative models have seen significant advancements in recent years, yet often remain challenging and costly to train and use. We introduce Generative Topological Networks (GTNs) -- a new class of generative models that addresses these shortcomings. GTNs are trained deterministically using a simple supervised learning approach grounded in topology theory. GTNs are fast to train, and require only a single forward pass in a standard feedforward neural network to generate samples. We demonstrate the strengths of GTNs in several datasets, including MNIST, celebA and the Hands and Palm Images dataset. Finally, the theory behind GTNs offers insights into how to train generative models for improved performance.
Authors: Steve Azzolin, Antonio Longa, Stefano Teso, Andrea Passerini
Abstract: As Graph Neural Networks (GNNs) become more pervasive, it becomes paramount to build robust tools for computing explanations of their predictions. A key desideratum is that these explanations are faithful, i.e., that they portray an accurate picture of the GNN's reasoning process. A number of different faithfulness metrics exist, begging the question of what faithfulness is exactly, and what its properties are. We begin by showing that existing metrics are not interchangeable -- i.e., explanations attaining high faithfulness according to one metric may be unfaithful according to others -- and can be systematically insensitive to important properties of the explanation, and suggest how to address these issues. We proceed to show that, surprisingly, optimizing for faithfulness is not always a sensible design goal. Specifically, we show that for injective regular GNN architectures, perfectly faithful explanations are completely uninformative. The situation is different for modular GNNs, such as self-explainable and domain-invariant architectures, where optimizing faithfulness does not compromise informativeness, and is also unexpectedly tied to out-of-distribution generalization.
Authors: Petr Ry\v{s}av\'y, Xiaoyu He, Jakub Mare\v{c}ek
Abstract: We present a benchmark for methods in causal learning. Specifically, we consider training a rich class of causal models from time-series data, and we suggest the use of the Krebs cycle and models of metabolism more broadly.
Authors: Ali Naseh, Jaechul Roh, Eugene Bagdasaryan, Amir Houmansadr
Abstract: Recent advances in large text-conditional image generative models such as Stable Diffusion, Midjourney, and DALL-E 3 have revolutionized the field of image generation, allowing users to produce high-quality, realistic images from textual prompts. While these developments have enhanced artistic creation and visual communication, they also present an underexplored attack opportunity: the possibility of inducing biases by an adversary into the generated images for malicious intentions, e.g., to influence society and spread propaganda. In this paper, we demonstrate the possibility of such a bias injection threat by an adversary who backdoors such models with a small number of malicious data samples; the implemented backdoor is activated when special triggers exist in the input prompt of the backdoored models. On the other hand, the model's utility is preserved in the absence of the triggers, making the attack highly undetectable. We present a novel framework that enables efficient generation of poisoning samples with composite (multi-word) triggers for such an attack. Our extensive experiments using over 1 million generated images and against hundreds of fine-tuned models demonstrate the feasibility of the presented backdoor attack. We illustrate how these biases can bypass conventional detection mechanisms, highlighting the challenges in proving the existence of biases within operational constraints. Our cost analysis confirms the low financial barrier to executing such attacks, underscoring the need for robust defensive strategies against such vulnerabilities in text-to-image generation models.
Authors: Pavel Ryt\'i\v{r}, Ale\v{s} Wodecki, Jakub Mare\v{c}ek
Abstract: There has been a growing interest in causal learning in recent years. Commonly used representations of causal structures, including Bayesian networks and structural equation models (SEM), take the form of directed acyclic graphs (DAGs). We provide a novel mixed-integer quadratic programming formulation and associated algorithm that identifies DAGs on up to 50 vertices, where these are identifiable. We call this method ExDAG, which stands for Exact learning of DAGs. Although there is a superexponential number of constraints that prevent the formation of cycles, the algorithm adds constraints violated by solutions found, rather than imposing all constraints in each continuous-valued relaxation. Our empirical results show that ExDAG outperforms local state-of-the-art solvers in terms of precision and outperforms state-of-the-art global solvers with respect to scaling, when considering Gaussian noise. We also provide validation with respect to other noise distributions.
Authors: Yuxing Liu, Rui Pan, Tong Zhang
Abstract: Adaptive gradient algorithms have been widely adopted in training large-scale deep neural networks, especially large foundation models. Despite their huge success in practice, their theoretical advantages over stochastic gradient descent (SGD) have not been fully understood, especially in the large batch-size setting commonly used in practice. This is because the only theoretical result that can demonstrate the benefit of Adagrad over SGD was obtained in the original paper of Adagrad for nonsmooth objective functions. However, for nonsmooth objective functions, there can be a linear slowdown of convergence when batch size increases, and thus a convergence analysis based on nonsmooth assumption cannot be used for large batch algorithms. In this work, we resolve this gap between theory and practice by providing a new analysis of Adagrad on both convex and nonconvex smooth objectives suitable for the large batch setting. It is shown that under the anisotropic smoothness and noise conditions, increased batch size does not slow down convergence for Adagrad, and thus it can still achieve a faster convergence guarantee over SGD even in the large batch setting. We present detailed comparisons between SGD and Adagrad to provide a better understanding of the benefits of adaptive gradient methods. Experiments in logistic regression and instruction following fine-tuning tasks provide strong evidence to support our theoretical analysis.
Authors: Sattar Vakili
Abstract: Reinforcement Learning (RL) has shown great empirical success in various application domains. The theoretical aspects of the problem have been extensively studied over past decades, particularly under tabular and linear Markov Decision Process structures. Recently, non-linear function approximation using kernel-based prediction has gained traction. This approach is particularly interesting as it naturally extends the linear structure, and helps explain the behavior of neural-network-based models at their infinite width limit. The analytical results however do not adequately address the performance guarantees for this case. We will highlight this open problem, overview existing partial results, and discuss related challenges.
Authors: Austin Coursey, Junyi Ji, Marcos Quinones-Grueiro, William Barbour, Yuhang Zhang, Tyler Derr, Gautam Biswas
Abstract: Early and accurate detection of anomalous events on the freeway, such as accidents, can improve emergency response and clearance. However, existing delays and errors in event identification and reporting make it a difficult problem to solve. Current large-scale freeway traffic datasets are not designed for anomaly detection and ignore these challenges. In this paper, we introduce the first large-scale lane-level freeway traffic dataset for anomaly detection. Our dataset consists of a month of weekday radar detection sensor data collected in 4 lanes along an 18-mile stretch of Interstate 24 heading toward Nashville, TN, comprising over 3.7 million sensor measurements. We also collect official crash reports from the Nashville Traffic Management Center and manually label all other potential anomalies in the dataset. To show the potential for our dataset to be used in future machine learning and traffic research, we benchmark numerous deep learning anomaly detection models on our dataset. We find that unsupervised graph neural network autoencoders are a promising solution for this problem and that ignoring spatial relationships leads to decreased performance. We demonstrate that our methods can reduce reporting delays by over 10 minutes on average while detecting 75% of crashes. Our dataset and all preprocessing code needed to get started are publicly released at https://vu.edu/ft-aed/ to facilitate future research.
URLs: https://vu.edu/ft-aed/
Authors: Amanda A. Volk, Kristofer G. Reyes, Jeffrey G. Ethier, Luke A. Baldwin
Abstract: Asynchronous Bayesian optimization is a recently implemented technique that allows for parallel operation of experimental systems and disjointed workflows. Contrasting with serial Bayesian optimization which individually selects experiments one at a time after conducting a measurement for each experiment, asynchronous policies sequentially assign multiple experiments before measurements can be taken and evaluate new measurements continuously as they are made available. This technique allows for faster data generation and therefore faster optimization of an experimental space. This work extends the capabilities of asynchronous optimization methods beyond prior studies by evaluating four additional policies that incorporate pessimistic predictions in the training data set. Combined with a conventional greedy policy, the five total policies were evaluated in a simulated environment and benchmarked with serial sampling. Under some conditions and parameter space dimensionalities, the pessimistic asynchronous policy reached optimum experimental conditions in significantly fewer experiments than equivalent serial policies and proved to be less susceptible to convergence onto local optima at higher dimensions. Without accounting for the faster sampling rate, the pessimistic asynchronous algorithm presented in this work could result in more efficient algorithm driven optimization of high-cost experimental spaces. Accounting for sampling rate, the presented asynchronous algorithm could allow for faster optimization in experimental spaces where multiple experiments can be run before results are collected.
Authors: Zesheng Liu, Maryam Rahnemoonfar
Abstract: Learning spatio-temporal patterns of polar ice layers is crucial for monitoring the change in ice sheet balance and evaluating ice dynamic processes. While a few researchers focus on learning ice layer patterns from echogram images captured by airborne snow radar sensors via different convolutional neural networks, the noise in the echogram images proves to be a major obstacle. Instead, we focus on geometric deep learning based on graph neural networks to learn the spatio-temporal patterns from thickness information of shallow ice layers and make predictions for deep layers. In this paper, we propose a physics-informed hybrid graph neural network that combines the GraphSAGE framework for graph feature learning with the long short-term memory (LSTM) structure for learning temporal changes, and introduce measurements of physical ice properties from Model Atmospheric Regional (MAR) weather model as physical node features. We found that our proposed network can consistently outperform the current non-inductive or non-physical model in predicting deep ice layer thickness.
Authors: Jinyin Wang, Haijing Zhang, Yihao Zhong, Yingbin Liang, Rongwei Ji, Yiru Cang
Abstract: Image-text matching is a key multimodal task that aims to model the semantic association between images and text as a matching relationship. With the advent of the multimedia information age, image, and text data show explosive growth, and how to accurately realize the efficient and accurate semantic correspondence between them has become the core issue of common concern in academia and industry. In this study, we delve into the limitations of current multimodal deep learning models in processing image-text pairing tasks. Therefore, we innovatively design an advanced multimodal deep learning architecture, which combines the high-level abstract representation ability of deep neural networks for visual information with the advantages of natural language processing models for text semantic understanding. By introducing a novel cross-modal attention mechanism and hierarchical feature fusion strategy, the model achieves deep fusion and two-way interaction between image and text feature space. In addition, we also optimize the training objectives and loss functions to ensure that the model can better map the potential association structure between images and text during the learning process. Experiments show that compared with existing image-text matching models, the optimized new model has significantly improved performance on a series of benchmark data sets. In addition, the new model also shows excellent generalization and robustness on large and diverse open scenario datasets and can maintain high matching performance even in the face of previously unseen complex situations.
Authors: Raphael Azorin, Zied Ben Houidi, Massimo Gallo, Alessandro Finamore, Pietro Michiardi
Abstract: Tabular data is ubiquitous in many real-life systems. In particular, time-dependent tabular data, where rows are chronologically related, is typically used for recording historical events, e.g., financial transactions, healthcare records, or stock history. Recently, hierarchical variants of the attention mechanism of transformer architectures have been used to model tabular time-series data. At first, rows (or columns) are encoded separately by computing attention between their fields. Subsequently, encoded rows (or columns) are attended to one another to model the entire tabular time-series. While efficient, this approach constrains the attention granularity and limits its ability to learn patterns at the field-level across separate rows, or columns. We take a first step to address this gap by proposing Fieldy, a fine-grained hierarchical model that contextualizes fields at both the row and column levels. We compare our proposal against state of the art models on regression and classification tasks using public tabular time-series datasets. Our results show that combining row-wise and column-wise attention improves performance without increasing model size. Code and data are available at https://github.com/raphaaal/fieldy.
Authors: Haoyang Liu, Haohan Wang
Abstract: Recent advancements in machine learning have significantly improved the identification of disease-associated genes from gene expression datasets. However, these processes often require extensive expertise and manual effort, limiting their scalability. Large Language Model (LLM)-based agents have shown promise in automating these tasks due to their increasing problem-solving abilities. To support the evaluation and development of such methods, we introduce GenoTEX, a benchmark dataset for the automatic exploration of gene expression data, involving the tasks of dataset selection, preprocessing, and statistical analysis. GenoTEX provides annotated code and results for solving a wide range of gene identification problems, in a full analysis pipeline that follows the standard of computational genomics. These annotations are curated by human bioinformaticians who carefully analyze the datasets to ensure accuracy and reliability. To provide baselines for these tasks, we present GenoAgents, a team of LLM-based agents designed with context-aware planning, iterative correction, and domain expert consultation to collaboratively explore gene datasets. Our experiments with GenoAgents demonstrate the potential of LLM-based approaches in genomics data analysis, while error analysis highlights the challenges and areas for future improvement. We propose GenoTEX as a promising resource for benchmarking and enhancing AI-driven methods for genomics data analysis. We make our benchmark publicly available at \url{https://github.com/Liu-Hy/GenoTex}.
Authors: Chengzhe Piao, Taiyu Zhu, Yu Wang, Stephanie E Baldeweg, Paul Taylor, Pantelis Georgiou, Jiahao Sun, Jun Wang, Kezhi Li
Abstract: Newly diagnosed Type 1 Diabetes (T1D) patients often struggle to obtain effective Blood Glucose (BG) prediction models due to the lack of sufficient BG data from Continuous Glucose Monitoring (CGM), presenting a significant "cold start" problem in patient care. Utilizing population models to address this challenge is a potential solution, but collecting patient data for training population models in a privacy-conscious manner is challenging, especially given that such data is often stored on personal devices. Considering the privacy protection and addressing the "cold start" problem in diabetes care, we propose "GluADFL", blood Glucose prediction by Asynchronous Decentralized Federated Learning. We compared GluADFL with eight baseline methods using four distinct T1D datasets, comprising 298 participants, which demonstrated its superior performance in accurately predicting BG levels for cross-patient analysis. Furthermore, patients' data might be stored and shared across various communication networks in GluADFL, ranging from highly interconnected (e.g., random, performs the best among others) to more structured topologies (e.g., cluster and ring), suitable for various social networks. The asynchronous training framework supports flexible participation. By adjusting the ratios of inactive participants, we found it remains stable if less than 70% are inactive. Our results confirm that GluADFL offers a practical, privacy-preserving solution for BG prediction in T1D, significantly enhancing the quality of diabetes management.
Authors: Yunjae Lee, Hyeseong Kim, Minsoo Rhu
Abstract: Training recommendation systems (RecSys) faces several challenges as it requires the "data preprocessing" stage to preprocess an ample amount of raw data and feed them to the GPU for training in a seamless manner. To sustain high training throughput, state-of-the-art solutions reserve a large fleet of CPU servers for preprocessing which incurs substantial deployment cost and power consumption. Our characterization reveals that prior CPU-centric preprocessing is bottlenecked on feature generation and feature normalization operations as it fails to reap out the abundant inter-/intra-feature parallelism in RecSys preprocessing. PreSto is a storage-centric preprocessing system leveraging In-Storage Processing (ISP), which offloads the bottlenecked preprocessing operations to our ISP units. We show that PreSto outperforms the baseline CPU-centric system with a $9.6\times$ speedup in end-to-end preprocessing time, $4.3\times$ enhancement in cost-efficiency, and $11.3\times$ improvement in energyefficiency on average for production-scale RecSys preprocessing.
Authors: Hyunjun Ma, Jin-soo Kim, Jong-Ho Choe, Q-Han Park
Abstract: We propose a metasurface antenna capable of real time holographic beam steering. An array of reconfigurable dipoeles can generate on demand far field patterns of radiation through the specific encoding of meta atomic states. i.e., the configuration of each dipole. Suitable states for the generation of the desired patterns can be identified using iteartion, but this is very slow and needs to be done for each far field pattern. Here, we present a deep learning based method for the control of a metasurface antenna with point dipole elements that vary in their state using dipole polarizability. Instead of iteration, we adopt a deep learning algorithm that combines an autoencoder with an electromagnetic scattering equation to determin the states required for a target far field pattern in real time. The scattering equation from Born approximation is used as the decoder in training the neural network, and analytic Green's function calculation is used to check the validity of Born approximation. Our learning based algorithm requires a computing time of within in 200 microseconds to determine the meta atomic states, thus enabling the real time opeartion of a holographic antenna.
Authors: Erik Jones, Anca Dragan, Jacob Steinhardt
Abstract: Developers try to evaluate whether an AI system can be misused by adversaries before releasing it; for example, they might test whether a model enables cyberoffense, user manipulation, or bioterrorism. In this work, we show that individually testing models for misuse is inadequate; adversaries can misuse combinations of models even when each individual model is safe. The adversary accomplishes this by first decomposing tasks into subtasks, then solving each subtask with the best-suited model. For example, an adversary might solve challenging-but-benign subtasks with an aligned frontier model, and easy-but-malicious subtasks with a weaker misaligned model. We study two decomposition methods: manual decomposition where a human identifies a natural decomposition of a task, and automated decomposition where a weak model generates benign tasks for a frontier model to solve, then uses the solutions in-context to solve the original task. Using these decompositions, we empirically show that adversaries can create vulnerable code, explicit images, python scripts for hacking, and manipulative tweets at much higher rates with combinations of models than either individual model. Our work suggests that even perfectly-aligned frontier systems can enable misuse without ever producing malicious outputs, and that red-teaming efforts should extend beyond single models in isolation.
Authors: Gabriel Sarch, Lawrence Jang, Michael J. Tarr, William W. Cohen, Kenneth Marino, Katerina Fragkiadaki
Abstract: Large-scale generative language and vision-language models (LLMs and VLMs) excel in few-shot in-context learning for decision making and instruction following. However, they require high-quality exemplar demonstrations to be included in their context window. In this work, we ask: Can LLMs and VLMs generate their own prompt examples from generic, sub-optimal demonstrations? We propose In-Context Abstraction Learning (ICAL), a method that builds a memory of multimodal experience insights from sub-optimal demonstrations and human feedback. Given a noisy demonstration in a new domain, VLMs abstract the trajectory into a general program by fixing inefficient actions and annotating cognitive abstractions: task relationships, object state changes, temporal subgoals, and task construals. These abstractions are refined and adapted interactively through human feedback while the agent attempts to execute the trajectory in a similar environment. The resulting abstractions, when used as exemplars in the prompt, significantly improve decision-making in retrieval-augmented LLM and VLM agents. Our ICAL agent surpasses the state-of-the-art in dialogue-based instruction following in TEACh, multimodal web agents in VisualWebArena, and action anticipation in Ego4D. In TEACh, we achieve a 12.6% improvement in goal-condition success. In VisualWebArena, our task success rate improves over the SOTA from 14.3% to 22.7%. In Ego4D action forecasting, we improve over few-shot GPT-4V and remain competitive with supervised models. We show finetuning our retrieval-augmented in-context agent yields additional improvements. Our approach significantly reduces reliance on expert-crafted examples and consistently outperforms in-context learning from action plans that lack such insights.
Authors: Yile Liang, Jiuxia Zhao, Donghui Li, Jie Feng, Chen Zhang, Xuetao Ding, Jinghua Hao, Renqing He
Abstract: The recent past has witnessed a notable surge in on-demand food delivery (OFD) services, offering delivery fulfillment within dozens of minutes after an order is placed. In OFD, pooling multiple orders for simultaneous delivery in real-time order assignment is a pivotal efficiency source, which may in turn extend delivery time. Constructing high-quality order pooling to harmonize platform efficiency with the experiences of consumers and couriers, is crucial to OFD platforms. However, the complexity and real-time nature of order assignment, making extensive calculations impractical, significantly limit the potential for order consolidation. Moreover, offline environment is frequently riddled with unknown factors, posing challenges for the platform's perceptibility and pooling decisions. Nevertheless, delivery behaviors of skilled couriers (SCs) who know the environment well, can improve system awareness and effectively inform decisions. Hence a SC delivery network (SCDN) is constructed, based on an enhanced attributed heterogeneous network embedding approach tailored for OFD. It aims to extract features from rich temporal and spatial information, and uncover the latent potential for order combinations embedded within SC trajectories. Accordingly, the vast search space of order assignment can be effectively pruned through scalable similarity calculations of low-dimensional vectors, making comprehensive and high-quality pooling outcomes more easily identified in real time. SCDN has now been deployed in Meituan dispatch system. Online tests reveal that with SCDN, the pooling quality and extent have been greatly improved. And our system can boost couriers'efficiency by 45-55% during noon peak hours, while upholding the timely delivery commitment.
Authors: Kawshik Manikantan (CVIT, IIIT Hyderabad), Shubham Toshniwal (NVIDIA), Makarand Tapaswi (CVIT, IIIT Hyderabad), Vineet Gandhi (CVIT, IIIT Hyderabad)
Abstract: The limited generalization of coreference resolution (CR) models has been a major bottleneck in the task's broad application. Prior work has identified annotation differences, especially for mention detection, as one of the main reasons for the generalization gap and proposed using additional annotated target domain data. Rather than relying on this additional annotation, we propose an alternative formulation of the CR task, Major Entity Identification (MEI), where we: (a) assume the target entities to be specified in the input, and (b) limit the task to only the frequent entities. Through extensive experiments, we demonstrate that MEI models generalize well across domains on multiple datasets with supervised models and LLM-based few-shot prompting. Additionally, the MEI task fits the classification framework, which enables the use of classification-based metrics that are more robust than the current CR metrics. Finally, MEI is also of practical use as it allows a user to search for all mentions of a particular entity or a group of entities of interest.
Authors: Jin Wang, Rui Dai, Weijie Wang, Luca Rossini, Francesco Ruscelli, Nikos Tsagarakis
Abstract: Enabling robots to autonomously perform hybrid motions in diverse environments can be beneficial for long-horizon tasks such as material handling, household chores, and work assistance. This requires extensive exploitation of intrinsic motion capabilities, extraction of affordances from rich environmental information, and planning of physical interaction behaviors. Despite recent progress has demonstrated impressive humanoid whole-body control abilities, they struggle to achieve versatility and adaptability for new tasks. In this work, we propose HYPERmotion, a framework that learns, selects and plans behaviors based on tasks in different scenarios. We combine reinforcement learning with whole-body optimization to generate motion for 38 actuated joints and create a motion library to store the learned skills. We apply the planning and reasoning features of the large language models (LLMs) to complex loco-manipulation tasks, constructing a hierarchical task graph that comprises a series of primitive behaviors to bridge lower-level execution with higher-level planning. By leveraging the interaction of distilled spatial geometry and 2D observation with a visual language model (VLM) to ground knowledge into a robotic morphology selector to choose appropriate actions in single- or dual-arm, legged or wheeled locomotion. Experiments in simulation and real-world show that learned motions can efficiently adapt to new tasks, demonstrating high autonomy from free-text commands in unstructured scenes. Videos and website: hy-motion.github.io/
Authors: Allen Roush, Yusuf Shabazz, Arvind Balaji, Peter Zhang, Stefano Mezza, Markus Zhang, Sanjay Basu, Sriram Vishwanath, Mehdi Fatemi, Ravid Schwartz-Ziv
Abstract: We introduce OpenDebateEvidence, a comprehensive dataset for argument mining and summarization sourced from the American Competitive Debate community. This dataset includes over 3.5 million documents with rich metadata, making it one of the most extensive collections of debate evidence. OpenDebateEvidence captures the complexity of arguments in high school and college debates, providing valuable resources for training and evaluation. Our extensive experiments demonstrate the efficacy of fine-tuning state-of-the-art large language models for argumentative abstractive summarization across various methods, models, and datasets. By providing this comprehensive resource, we aim to advance computational argumentation and support practical applications for debaters, educators, and researchers. OpenDebateEvidence is publicly available to support further research and innovation in computational argumentation. Access it here: https://huggingface.co/datasets/Yusuf5/OpenCaselist
Authors: Atula Tejaswi, Nilesh Gupta, Eunsol Choi
Abstract: Despite rapid progress in large language models (LLMs), their performance on a vast majority of languages remain unsatisfactory. In this paper, we study building language-specific LLMs by adapting monolingual and multilingual LLMs. We conduct systematic experiments on how design choices (base model selection, vocabulary extension, and continued fine-tuning) impact the adapted LLM, both in terms of efficiency (how many tokens are needed to encode the same amount of information) and end task performance. We find that (1) the initial performance before the adaptation is not always indicative of the final performance. (2) Efficiency can easily improved with simple vocabulary extension and continued fine-tuning in most LLMs we study, and (3) The optimal adaptation method is highly language-dependent, and the simplest approach works well across various experimental settings. Adapting English-centric models can yield better results than adapting multilingual models despite their worse initial performance on low-resource languages. Together, our work lays foundations on efficiently building language-specific LLMs by adapting existing LLMs.
Authors: Frank Willard, Luke Moffett, Emmanuel Mokel, Jon Donnelly, Stark Guo, Julia Yang, Giyoung Kim, Alina Jade Barnett, Cynthia Rudin
Abstract: Prototypical-part models are a popular interpretable alternative to black-box deep learning models for computer vision. However, they are difficult to train, with high sensitivity to hyperparameter tuning, inhibiting their application to new datasets and our understanding of which methods truly improve their performance. To facilitate the careful study of prototypical-part networks (ProtoPNets), we create a new framework for integrating components of prototypical-part models -- ProtoPNeXt. Using ProtoPNeXt, we show that applying Bayesian hyperparameter tuning and an angular prototype similarity metric to the original ProtoPNet is sufficient to produce new state-of-the-art accuracy for prototypical-part models on CUB-200 across multiple backbones. We further deploy this framework to jointly optimize for accuracy and prototype interpretability as measured by metrics included in ProtoPNeXt. Using the same resources, this produces models with substantially superior semantics and changes in accuracy between +1.3% and -1.5%. The code and trained models will be made publicly available upon publication.
Authors: Rachel Morris, Ryan Murray
Abstract: In recent years there has been significant interest in the effect of different types of adversarial perturbations in data classification problems. Many of these models incorporate the adversarial power, which is an important parameter with an associated trade-off between accuracy and robustness. This work considers a general framework for adversarially-perturbed classification problems, in a large data or population-level limit. In such a regime, we demonstrate that as adversarial strength goes to zero that optimal classifiers converge to the Bayes classifier in the Hausdorff distance. This significantly strengthens previous results, which generally focus on $L^1$-type convergence. The main argument relies upon direct geometric comparisons and is inspired by techniques from geometric measure theory.
Authors: Alkis Koudounas, Flavio Giobergia, Eliana Pastor, Elena Baralis
Abstract: Speech models may be affected by performance imbalance in different population subgroups, raising concerns about fair treatment across these groups. Prior attempts to mitigate unfairness either focus on user-defined subgroups, potentially overlooking other affected subgroups, or do not explicitly improve the internal representation at the subgroup level. This paper proposes the first adoption of contrastive learning to mitigate speech model bias in underperforming subgroups. We employ a three-level learning technique that guides the model in focusing on different scopes for the contrastive loss, i.e., task, subgroup, and the errors within subgroups. The experiments on two spoken language understanding datasets and two languages demonstrate that our approach improves internal subgroup representations, thus reducing model bias and enhancing performance.
Authors: Alkis Koudounas, Gabriele Ciravegna, Marco Fantini, Giovanni Succo, Erika Crosetti, Tania Cerquitelli, Elena Baralis
Abstract: Voice disorders are pathologies significantly affecting patient quality of life. However, non-invasive automated diagnosis of these pathologies is still under-explored, due to both a shortage of pathological voice data, and diversity of the recording types used for the diagnosis. This paper proposes a novel solution that adopts transformers directly working on raw voice signals and addresses data shortage through synthetic data generation and data augmentation. Further, we consider many recording types at the same time, such as sentence reading and sustained vowel emission, by employing a Mixture of Expert ensemble to align the predictions on different data types. The experimental results, obtained on both public and private datasets, show the effectiveness of our solution in the disorder detection and classification tasks and largely improve over existing approaches.
Authors: Maximilian Wiesmann, Paul Larsen
Abstract: In this paper we demonstrate the possibility of a trend reversal in binary classification tasks between the dataset and a classification score obtained from a trained model. This trend reversal occurs for certain choices of the regularization parameter for model training, namely, if the parameter is contained in what we call the pathological regularization regime. For ridge regression, we give necessary and sufficient algebraic conditions on the dataset for the existence of a pathological regularization regime. Moreover, our results provide a data science practitioner with a hands-on tool to avoid hyperparameter choices suffering from trend reversal. We furthermore present numerical results on pathological regularization regimes for logistic regression. Finally, we draw connections to datasets exhibiting Simpson's paradox, providing a natural source of pathological datasets.
Authors: Lexin Zhou, Youmna Farag, Andreas Vlachos
Abstract: Research on dialogue constructiveness assessment focuses on (i) analysing conversational factors that influence individuals to take specific actions, win debates, change their perspectives or broaden their open-mindedness and (ii) predicting constructive outcomes following dialogues for such use cases. These objectives can be achieved by training either interpretable feature-based models (which often involve costly human annotations) or neural models such as pre-trained language models (which have empirically shown higher task accuracy but lack interpretability). We propose a novel LLM feature-based framework that combines the strengths of feature-based and neural approaches while mitigating their downsides, in assessing dialogue constructiveness. The framework first defines a set of dataset-independent and interpretable linguistic features, which can be extracted by both prompting an LLM and simple heuristics. Such features are then used to train LLM feature-based models. We apply this framework to three datasets of dialogue constructiveness and find that our LLM feature-based models significantly outperform standard feature-based models and neural models, and tend to learn more robust prediction rules instead of relying on superficial shortcuts (as seen with neural models). Further, we demonstrate that interpreting these LLM feature-based models can yield valuable insights into what makes a dialogue constructive.
Authors: Denis Rakitin, Ivan Shchekotov, Dmitry Vetrov
Abstract: Diffusion distillation methods aim to compress the diffusion models into efficient one-step generators while trying to preserve quality. Among them, Distribution Matching Distillation (DMD) offers a suitable framework for training general-form one-step generators, applicable beyond unconditional generation. In this work, we introduce its modification, called Regularized Distribution Matching Distillation, applicable to unpaired image-to-image (I2I) problems. We demonstrate its empirical performance in application to several translation tasks, including 2D examples and I2I between different image datasets, where it performs on par or better than multi-step diffusion baselines.
Authors: William Fleshman, Benjamin Van Durme
Abstract: Large language models (LLMs) fine-tuned for text-retrieval have demonstrated state-of-the-art results across several information retrieval (IR) benchmarks. However, supervised training for improving these models requires numerous labeled examples, which are generally unavailable or expensive to acquire. In this work, we explore the effectiveness of extending reverse engineered adaptation to the context of information retrieval (RE-AdaptIR). We use RE-AdaptIR to improve LLM-based IR models using only unlabeled data. We demonstrate improved performance both in training domains as well as zero-shot in domains where the models have seen no queries. We analyze performance changes in various fine-tuning scenarios and offer findings of immediate use to practitioners.
Authors: Steven A. Lehr, Aylin Caliskan, Suneragiri Liyanage, Mahzarin R. Banaji
Abstract: How good a research scientist is ChatGPT? We systematically probed the capabilities of GPT-3.5 and GPT-4 across four central components of the scientific process: as a Research Librarian, Research Ethicist, Data Generator, and Novel Data Predictor, using psychological science as a testing field. In Study 1 (Research Librarian), unlike human researchers, GPT-3.5 and GPT-4 hallucinated, authoritatively generating fictional references 36.0% and 5.4% of the time, respectively, although GPT-4 exhibited an evolving capacity to acknowledge its fictions. In Study 2 (Research Ethicist), GPT-4 (though not GPT-3.5) proved capable of detecting violations like p-hacking in fictional research protocols, correcting 88.6% of blatantly presented issues, and 72.6% of subtly presented issues. In Study 3 (Data Generator), both models consistently replicated patterns of cultural bias previously discovered in large language corpora, indicating that ChatGPT can simulate known results, an antecedent to usefulness for both data generation and skills like hypothesis generation. Contrastingly, in Study 4 (Novel Data Predictor), neither model was successful at predicting new results absent in their training data, and neither appeared to leverage substantially new information when predicting more versus less novel outcomes. Together, these results suggest that GPT is a flawed but rapidly improving librarian, a decent research ethicist already, capable of data generation in simple domains with known characteristics but poor at predicting novel patterns of empirical data to aid future experimentation.
Authors: Aman Gupta, Aditi Sheshadri, Sujit Roy, Vishal Gaur, Manil Maskey, Rahul Ramachandran
Abstract: Global climate models typically operate at a grid resolution of hundreds of kilometers and fail to resolve atmospheric mesoscale processes, e.g., clouds, precipitation, and gravity waves (GWs). Model representation of these processes and their sources is essential to the global circulation and planetary energy budget, but subgrid scale contributions from these processes are often only approximately represented in models using parameterizations. These parameterizations are subject to approximations and idealizations, which limit their capability and accuracy. The most drastic of these approximations is the "single-column approximation" which completely neglects the horizontal evolution of these processes, resulting in key biases in current climate models. With a focus on atmospheric GWs, we present the first-ever global simulation of atmospheric GW fluxes using machine learning (ML) models trained on the WINDSET dataset to emulate global GW emulation in the atmosphere, as an alternative to traditional single-column parameterizations. Using an Attention U-Net-based architecture trained on globally resolved GW momentum fluxes, we illustrate the importance and effectiveness of global nonlocality, when simulating GWs using data-driven schemes.
Authors: Gaurav Ghosal, Tatsunori Hashimoto, Aditi Raghunathan
Abstract: In this work, we study the impact of QA fine-tuning data on downstream factuality. We show that fine-tuning on lesser-known facts that are poorly stored during pretraining yields significantly worse factuality than fine-tuning on well-known facts, even when all facts are seen during pretraining. We prove this phenomenon theoretically, showing that training on lesser-known facts can lead the model to ignore subject entity names and instead output a generic plausible response even when the relevant factual knowledge is encoded in the model. On three question answering benchmarks (PopQA, Entity Questions, and MMLU) and two language models (Llama-2-7B and Mistral-7B), we find that (i) finetuning on a completely factual but lesser-known subset of the data deteriorates downstream factuality (5-10%) and (ii) finetuning on a subset of better-known examples matches or outperforms finetuning on the entire dataset. Ultimately, our results shed light on the interaction between pretrained knowledge and finetuning data and demonstrate the importance of taking into account how facts are stored in the pretrained model when fine-tuning for knowledge-intensive tasks.
Authors: Chen Liu, Ke Xu, Liangbo L. Shen, Guillaume Huguet, Zilong Wang, Alexander Tong, Danilo Bzdok, Jay Stewart, Jay C. Wang, Lucian V. Del Priore, Smita Krishnaswamy
Abstract: The forecasting of disease progression from images is a holy grail for clinical decision making. However, this task is complicated by the inherent high dimensionality, temporal sparsity and sampling irregularity in longitudinal image acquisitions. Existing methods often rely on extracting hand-crafted features and performing time-series analysis in this vector space, leading to a loss of rich spatial information within the images. To overcome these challenges, we introduce ImageFlowNet, a novel framework that learns latent-space flow fields that evolve multiscale representations in joint embedding spaces using neural ODEs and SDEs to model disease progression in the image domain. Notably, ImageFlowNet learns multiscale joint representation spaces by combining cohorts of patients together so that information can be transferred between the patient samples. The dynamics then provide plausible trajectories of progression, with the SDE providing alternative trajectories from the same starting point. We provide theoretical insights that support our formulation of ODEs, and motivate our regularizations involving high-level visual features, latent space organization, and trajectory smoothness. We then demonstrate ImageFlowNet's effectiveness through empirical evaluations on three longitudinal medical image datasets depicting progression in retinal geographic atrophy, multiple sclerosis, and glioblastoma.
Authors: Yi Sun, Debarghya Mukherjee, Yves Atchade
Abstract: Solving partial differential equations (PDEs) and their inverse problems using Physics-informed neural networks (PINNs) is a rapidly growing approach in the physics and machine learning community. Although several architectures exist for PINNs that work remarkably in practice, our theoretical understanding of their performances is somewhat limited. In this work, we study the behavior of a Bayesian PINN estimator of the solution of a PDE from $n$ independent noisy measurement of the solution. We focus on a class of equations that are linear in their parameters (with unknown coefficients $\theta_\star$). We show that when the partial differential equation admits a classical solution (say $u_\star$), differentiable to order $\beta$, the mean square error of the Bayesian posterior mean is at least of order $n^{-2\beta/(2\beta + d)}$. Furthermore, we establish a convergence rate of the linear coefficients of $\theta_\star$ depending on the order of the underlying differential operator. Last but not least, our theoretical results are validated through extensive simulations.
Authors: Guido Di Federico, Louis J. Durlofsky
Abstract: Geological parameterization entails the representation of a geomodel using a small set of latent variables and a mapping from these variables to grid-block properties such as porosity and permeability. Parameterization is useful for data assimilation (history matching), as it maintains geological realism while reducing the number of variables to be determined. Diffusion models are a new class of generative deep-learning procedures that have been shown to outperform previous methods, such as generative adversarial networks, for image generation tasks. Diffusion models are trained to "denoise", which enables them to generate new geological realizations from input fields characterized by random noise. Latent diffusion models, which are the specific variant considered in this study, provide dimension reduction through use of a low-dimensional latent variable. The model developed in this work includes a variational autoencoder for dimension reduction and a U-net for the denoising process. Our application involves conditional 2D three-facies (channel-levee-mud) systems. The latent diffusion model is shown to provide realizations that are visually consistent with samples from geomodeling software. Quantitative metrics involving spatial and flow-response statistics are evaluated, and general agreement between the diffusion-generated models and reference realizations is observed. Stability tests are performed to assess the smoothness of the parameterization method. The latent diffusion model is then used for ensemble-based data assimilation. Two synthetic "true" models are considered. Significant uncertainty reduction, posterior P$_{10}$-P$_{90}$ forecasts that generally bracket observed data, and consistent posterior geomodels, are achieved in both cases.
Authors: Tinh Son Luong, Thanh-Thien Le, Thang Viet Doan, Linh Ngo Van, Thien Huu Nguyen, Diep Thi-Ngoc Nguyen
Abstract: Existing toxic detection models face significant limitations, such as lack of transparency, customization, and reproducibility. These challenges stem from the closed-source nature of their training data and the paucity of explanations for their evaluation mechanism. To address these issues, we propose a dataset creation mechanism that integrates voting and chain-of-thought processes, producing a high-quality open-source dataset for toxic content detection. Our methodology ensures diverse classification metrics for each sample and includes both classification scores and explanatory reasoning for the classifications. We utilize the dataset created through our proposed mechanism to train our model, which is then compared against existing widely-used detectors. Our approach not only enhances transparency and customizability but also facilitates better fine-tuning for specific use cases. This work contributes a robust framework for developing toxic content detection models, emphasizing openness and adaptability, thus paving the way for more effective and user-specific content moderation solutions.
Authors: George D. Pasparakis, Lori Graham-Brady, Michael D. Shields
Abstract: Stress and material deformation field predictions are among the most important tasks in computational mechanics. These predictions are typically made by solving the governing equations of continuum mechanics using finite element analysis, which can become computationally prohibitive considering complex microstructures and material behaviors. Machine learning (ML) methods offer potentially cost effective surrogates for these applications. However, existing ML surrogates are either limited to low-dimensional problems and/or do not provide uncertainty estimates in the predictions. This work proposes an ML surrogate framework for stress field prediction and uncertainty quantification for diverse materials microstructures. A modified Bayesian U-net architecture is employed to provide a data-driven image-to-image mapping from initial microstructure to stress field with prediction (epistemic) uncertainty estimates. The Bayesian posterior distributions for the U-net parameters are estimated using three state-of-the-art inference algorithms: the posterior sampling-based Hamiltonian Monte Carlo method and two variational approaches, the Monte-Carlo Dropout method and the Bayes by Backprop algorithm. A systematic comparison of the predictive accuracy and uncertainty estimates for these methods is performed for a fiber reinforced composite material and polycrystalline microstructure application. It is shown that the proposed methods yield predictions of high accuracy compared to the FEA solution, while uncertainty estimates depend on the inference approach. Generally, the Hamiltonian Monte Carlo and Bayes by Backprop methods provide consistent uncertainty estimates. Uncertainty estimates from Monte Carlo Dropout, on the other hand, are more difficult to interpret and depend strongly on the method's design.
Authors: Yihao Zheng, Haocheng Xia, Junyuan Pang, Jinfei Liu, Kui Ren, Lingyang Chu, Yang Cao, Li Xiong
Abstract: Watermarking is broadly utilized to protect ownership of shared data while preserving data utility. However, existing watermarking methods for tabular datasets fall short on the desired properties (detectability, non-intrusiveness, and robustness) and only preserve data utility from the perspective of data statistics, ignoring the performance of downstream ML models trained on the datasets. Can we watermark tabular datasets without significantly compromising their utility for training ML models while preventing attackers from training usable ML models on attacked datasets? In this paper, we propose a hypothesis testing-based watermarking scheme, TabularMark. Data noise partitioning is utilized for data perturbation during embedding, which is adaptable for numerical and categorical attributes while preserving the data utility. For detection, a custom-threshold one proportion z-test is employed, which can reliably determine the presence of the watermark. Experiments on real-world and synthetic datasets demonstrate the superiority of TabularMark in detectability, non-intrusiveness, and robustness.
Authors: Md Saiful Islam, Tariq Adnan, Jan Freyberg, Sangwu Lee, Abdelrahman Abdelkader, Meghan Pawlik, Cathe Schwartz, Karen Jaffe, Ruth B. Schneider, E Ray Dorsey, Ehsan Hoque
Abstract: Limited access to neurological care leads to missed diagnoses of Parkinson's disease (PD), leaving many individuals unidentified and untreated. We trained a novel neural network-based fusion architecture to detect Parkinson's disease (PD) by analyzing features extracted from webcam recordings of three tasks: finger tapping, facial expression (smiling), and speech (uttering a sentence containing all letters of the alphabet). Additionally, the model incorporated Monte Carlo Dropout to improve prediction accuracy by considering uncertainties. The study participants (n = 845, 272 with PD) were randomly split into three sets: 60% for training, 20% for model selection (hyper-parameter tuning), and 20% for final performance evaluation. The dataset consists of 1102 sessions, each session containing videos of all three tasks. Our proposed model achieved significantly better accuracy, area under the ROC curve (AUROC), and sensitivity at non-inferior specificity compared to any single-task model. Withholding uncertain predictions further boosted the performance, achieving 88.0% (95% CI: 87.7% - 88.4%) accuracy, 93.0% (92.8% - 93.2%) AUROC, 79.3% (78.4% - 80.2%) sensitivity, and 92.6% (92.3% - 92.8%) specificity, at the expense of not being able to predict for 2.3% (2.0% - 2.6%) data. Further analysis suggests that the trained model does not exhibit any detectable bias across sex and ethnic subgroups and is most effective for individuals aged between 50 and 80. This accessible, low-cost approach requiring only an internet-enabled device with a webcam and microphone paves the way for convenient PD screening at home, particularly in regions with limited access to clinical specialists.
Authors: Wentao Shi, Mengqi Yuan, Junkang Wu, Qifan Wang, Fuli Feng
Abstract: Adapting Large Language Models (LLMs) for agent tasks is critical in developing language agents. Direct Preference Optimization (DPO) is a promising technique for this adaptation with the alleviation of compounding errors, offering a means to directly optimize Reinforcement Learning (RL) objectives. However, applying DPO to multi-turn tasks presents challenges due to the inability to cancel the partition function. Overcoming this obstacle involves making the partition function independent of the current state and addressing length disparities between preferred and dis-preferred trajectories. In this light, we replace the policy constraint with the state-action occupancy measure constraint in the RL objective and add length normalization to the Bradley-Terry model, yielding a novel loss function named DMPO for multi-turn agent tasks with theoretical explanations. Extensive experiments on three multi-turn agent task datasets confirm the effectiveness and superiority of the DMPO loss.
Authors: Simon Kloker, Matthew Bazanya, Twaha Kateete
Abstract: Trust plays a pivotal role in Lecturer-Student-Collaboration, encompassing teaching and research aspects. The advent of Large Language Models (LLMs) in platforms like Open AI's ChatGPT, coupled with their cost-effectiveness and high-quality results, has led to their rapid adoption among university students. However, discerning genuine student input from LLM-generated output poses a challenge for lecturers. This dilemma jeopardizes the trust relationship between lecturers and students, potentially impacting university downstream activities, particularly collaborative research initiatives. Despite attempts to establish guidelines for student LLM use, a clear framework mutually beneficial for lecturers and students in higher education remains elusive. This study addresses the research question: How does the use of LLMs by students impact Informational and Procedural Justice, influencing Team Trust and Expected Team Performance? Methodically, we applied a quantitative construct-based survey, evaluated using techniques of Structural Equation Modelling (PLS- SEM) to examine potential relationships among these constructs. Our findings based on 23 valid respondents from Ndejje University indicate that lecturers are less concerned about the fairness of LLM use per se but are more focused on the transparency of student utilization, which significantly influences Team Trust positively. This research contributes to the global discourse on integrating and regulating LLMs and subsequent models in education. We propose that guidelines should support LLM use while enforcing transparency in Lecturer-Student- Collaboration to foster Team Trust and Performance. The study contributes valuable insights for shaping policies enabling ethical and transparent LLMs usage in education to ensure effectiveness of collaborative learning environments.
Authors: Zhuoxiao Chen, Junjie Meng, Mahsa Baktashmotlagh, Zi Huang, Yadan Luo
Abstract: LiDAR-based 3D object detection is pivotal across many applications, yet the performance of such detection systems often degrades after deployment, especially when faced with unseen test point clouds originating from diverse locations or subjected to corruption. In this work, we introduce a new online adaptation framework for detectors named Model Synergy (MOS). Specifically, MOS dynamically assembles best-fit supermodels for each test batch from a bank of historical checkpoints, leveraging long-term knowledge to guide model updates without forgetting. The model assembly is directed by the proposed synergy weights (SW), employed for weighted averaging of the selected checkpoints to minimize redundancy in the composite supermodel. These weights are calculated by evaluating the similarity of predicted bounding boxes on test data and the feature independence among model pairs in the bank. To maintain an informative yet compact model bank, we pop out checkpoints with the lowest average SW scores and insert newly updated model weights. Our method was rigorously tested against prior test-time domain adaptation strategies on three datasets and under eight types of corruptions, demonstrating its superior adaptability to changing scenes and conditions. Remarkably, our approach achieved a 67.3% increase in performance in a complex "cross-corruption" scenario, which involves cross-dataset inconsistencies and real-world scene corruptions, providing a more realistic testbed of adaptation capabilities. The code is available at https://github.com/zhuoxiao-chen/MOS.
Authors: Carlos Sebasti\'an, Carlos E. Gonz\'alez-Guill\'en, Jes\'us Juan
Abstract: Building prediction intervals for time series forecasting problems presents a complex challenge, particularly when relying solely on point predictors, a common scenario for practitioners in the industry. While research has primarily focused on achieving increasingly efficient valid intervals, we argue that, when evaluating a set of intervals, traditional measures alone are insufficient. There are additional crucial characteristics: the intervals must vary in length, with this variation directly linked to the difficulty of the prediction, and the coverage of the interval must remain independent of the difficulty of the prediction for practical utility. We propose the Heteroscedastic Quantile Regression (HQR) model and the Width-Adaptive Conformal Inference (WACI) method, providing theoretical coverage guarantees, to overcome those issues, respectively. The methodologies are evaluated in the context of Electricity Price Forecasting and Wind Power Forecasting, representing complex scenarios in time series forecasting. The results demonstrate that HQR and WACI not only improve or achieve typical measures of validity and efficiency but also successfully fulfil the commonly ignored mentioned characteristics.
Authors: Minjong Cheon
Abstract: In the realm of deep learning, the Kolmogorov-Arnold Network (KAN) has emerged as a potential alternative to multilayer projections (MLPs). However, its applicability to vision tasks has not been extensively validated. In our study, we demonstrated the effectiveness of KAN for vision tasks through multiple trials on the MNIST, CIFAR10, and CIFAR100 datasets, using a training batch size of 32. Our results showed that while KAN outperformed the original MLP-Mixer on CIFAR10 and CIFAR100, it performed slightly worse than the state-of-the-art ResNet-18. These findings suggest that KAN holds significant promise for vision tasks, and further modifications could enhance its performance in future evaluations.Our contributions are threefold: first, we showcase the efficiency of KAN-based algorithms for visual tasks; second, we provide extensive empirical assessments across various vision benchmarks, comparing KAN's performance with MLP-Mixer, CNNs, and Vision Transformers (ViT); and third, we pioneer the use of natural KAN layers in visual tasks, addressing a gap in previous research. This paper lays the foundation for future studies on KANs, highlighting their potential as a reliable alternative for image classification tasks.
Authors: Melvin Wong, Jiao Liu, Thiago Rios, Stefan Menzel, Yew Soon Ong
Abstract: The rapid research and development of generative artificial intelligence has enabled the generation of high-quality images, text, and 3D models from text prompts. This advancement impels an inquiry into whether these models can be leveraged to create digital artifacts for both creative and engineering applications. Drawing on innovative designs from other domains may be one answer to this question, much like the historical practice of ``bionics", where humans have sought inspiration from nature's exemplary designs. This raises the intriguing possibility of using generative models to simultaneously tackle design tasks across multiple domains, facilitating cross-domain learning and resulting in a series of innovative design solutions. In this paper, we propose LLM2FEA as the first attempt to discover novel designs in generative models by transferring knowledge across multiple domains. By utilizing a multi-factorial evolutionary algorithm (MFEA) to drive a large language model, LLM2FEA integrates knowledge from various fields to generate prompts that guide the generative model in discovering novel and practical objects. Experimental results in the context of 3D aerodynamic design verify the discovery capabilities of the proposed LLM2FEA. The designs generated by LLM2FEA not only satisfy practicality requirements to a certain degree but also feature novel and aesthetically pleasing shapes, demonstrating the potential applications of LLM2FEA in discovery tasks.
Authors: Guangkun Nie, Qinghao Zhao, Gongzheng Tang, Jun Li, Shenda Hong
Abstract: Photoplethysmography (PPG) is emerging as a crucial tool for monitoring human hemodynamics, with recent studies highlighting its potential in assessing vascular aging through deep learning. However, real-world age distributions are often imbalanced, posing significant challenges for deep learning models. In this paper, we introduce a novel, simple, and effective loss function named the Dist Loss to address deep imbalanced regression tasks. We trained a one-dimensional convolutional neural network (Net1D) incorporating the Dist Loss on the extensive UK Biobank dataset (n=502,389) to estimate vascular age from PPG signals and validate its efficacy in characterizing cardiovascular health. The model's performance was validated on a 40% held-out test set, achieving state-of-the-art results, especially in regions with small sample sizes. Furthermore, we divided the population into three subgroups based on the difference between predicted vascular age and chronological age: less than -10 years, between -10 and 10 years, and greater than 10 years. We analyzed the relationship between predicted vascular age and several cardiovascular events over a follow-up period of up to 10 years, including death, coronary heart disease, and heart failure. Our results indicate that the predicted vascular age has significant potential to reflect an individual's cardiovascular health status. Our code will be available at https://github.com/Ngk03/AI-vascular-age.
Authors: Shamane Siriwardhana, Mark McQuade, Thomas Gauthier, Lucas Atkins, Fernando Fernandes Neto, Luke Meyers, Anneketh Vij, Tyler Odenthal, Charles Goddard, Mary MacCarthy, Jacob Solawetz
Abstract: We conducted extensive experiments on domain adaptation of the Meta-Llama-3-70B-Instruct model on SEC data, exploring its performance on both general and domain-specific benchmarks. Our focus included continual pre-training (CPT) and model merging, aiming to enhance the model's domain-specific capabilities while mitigating catastrophic forgetting. Through this study, we evaluated the impact of integrating financial regulatory data into a robust language model and examined the effectiveness of our model merging techniques in preserving and improving the model's instructive abilities. The model is accessible at hugging face: https://huggingface.co/arcee-ai/Llama-3-SEC-Base, arcee-ai/Llama-3-SEC-Base. This is an intermediate checkpoint of our final model, which has seen 20B tokens so far. The full model is still in the process of training. This is a preprint technical report with thorough evaluations to understand the entire process.
Authors: Tatsuya Kamijo, Cristian C. Beltran-Hernandez, Masashi Hamaya
Abstract: Automating dexterous, contact-rich manipulation tasks using rigid robots is a significant challenge in robotics. Rigid robots, defined by their actuation through position commands, face issues of excessive contact forces due to their inability to adapt to contact with the environment, potentially causing damage. While compliance control schemes have been introduced to mitigate these issues by controlling forces via external sensors, they are hampered by the need for fine-tuning task-specific controller parameters. Learning from Demonstrations (LfD) offers an intuitive alternative, allowing robots to learn manipulations through observed actions. In this work, we introduce a novel system to enhance the teaching of dexterous, contact-rich manipulations to rigid robots. Our system is twofold: firstly, it incorporates a teleoperation interface utilizing Virtual Reality (VR) controllers, designed to provide an intuitive and cost-effective method for task demonstration with haptic feedback. Secondly, we present Comp-ACT (Compliance Control via Action Chunking with Transformers), a method that leverages the demonstrations to learn variable compliance control from a few demonstrations. Our methods have been validated across various complex contact-rich manipulation tasks using single-arm and bimanual robot setups in simulated and real-world environments, demonstrating the effectiveness of our system in teaching robots dexterous manipulations with enhanced adaptability and safety.
Authors: Benjamin Udofia, Tushar Jogi, Markus Stricker
Abstract: Detecting structure in data is the first step to arrive at meaningful representations for systems. This is particularly challenging for dislocation networks evolving as a consequence of plastic deformation of crystalline systems. Our study employs Isomap, a manifold learning technique, to unveil the intrinsic structure of high-dimensional density field data of dislocation structures from different compression axis. The resulting maps provide a systematic framework for quantitatively comparing dislocation structures, offering unique fingerprints based on density fields. Our novel, unbiased approach contributes to the quantitative classification of dislocation structures which can be systematically extended.
Authors: Aydin Abadi
Abstract: To securely transmit sensitive information into the future, Time-Lock Puzzles (TLPs) have been developed. Their applications include scheduled payments, timed commitments, e-voting, and sealed-bid auctions. Homomorphic TLP is a key variant of TLP that enables computation on puzzles from different clients. This allows a solver/server to tackle only a single puzzle encoding the computation's result. However, existing homomorphic TLPs lack support for verifying the correctness of the computation results. We address this limitation by introducing Tempora-Fusion, a TLP that allows a server to perform homomorphic linear combinations of puzzles from different clients while ensuring verification of computation correctness. This scheme avoids asymmetric-key cryptography for verification, thus paving the way for efficient implementations. We discuss our scheme's application in various domains, such as federated learning, scheduled payments in online banking, and e-voting.
Authors: Bram M. Renting, Thomas M. Moerland, Holger H. Hoos, Catholijn M. Jonker
Abstract: The research field of automated negotiation has a long history of designing agents that can negotiate with other agents. Such negotiation strategies are traditionally based on manual design and heuristics. More recently, reinforcement learning approaches have also been used to train agents to negotiate. However, negotiation problems are diverse, causing observation and action dimensions to change, which cannot be handled by default linear policy networks. Previous work on this topic has circumvented this issue either by fixing the negotiation problem, causing policies to be non-transferable between negotiation problems or by abstracting the observations and actions into fixed-size representations, causing loss of information and expressiveness due to feature design. We developed an end-to-end reinforcement learning method for diverse negotiation problems by representing observations and actions as a graph and applying graph neural networks in the policy. With empirical evaluations, we show that our method is effective and that we can learn to negotiate with other agents on never-before-seen negotiation problems. Our result opens up new opportunities for reinforcement learning in negotiation agents.
Authors: Seonggon Kim, Eunhyeok Park
Abstract: With the rapid increase in model size and the growing importance of various fine-tuning applications, lightweight training has become crucial. Since the backward pass is twice as expensive as the forward pass, optimizing backpropagation is particularly important. However, modifications to this process can lead to suboptimal convergence, so training optimization should minimize perturbations, which is a highly challenging task. In this study, we introduce a novel optimization strategy called Hadamard Low-rank Quantization (HLQ), focusing on reducing the cost of backpropagation in convolutional and linear layers. We first analyze the sensitivity of gradient computation with respect to activation and weight, and judiciously design the HLQ pipeline to apply 4-bit Hadamard quantization to the activation gradient and Hadamard low-rank approximation to the weight gradient. This combination was found to be the best for maximizing benefits, and our extensive experiments demonstrate the outstanding performance of HLQ in both training from scratch and fine-tuning, achieving significant memory savings and acceleration on real GPUs with negligible quality degradation.
Authors: Badr AlKhamissi, Greta Tuckute, Antoine Bosselut, Martin Schrimpf
Abstract: Large Language Models (LLMs) have been shown to be effective models of the human language system, with some models predicting most explainable variance of brain activity in current datasets. Even in untrained models, the representations induced by architectural priors can exhibit reasonable alignment to brain data. In this work, we investigate the key architectural components driving the surprising alignment of untrained models. To estimate LLM-to-brain similarity, we first select language-selective units within an LLM, similar to how neuroscientists identify the language network in the human brain. We then benchmark the brain alignment of these LLM units across five different brain recording datasets. By isolating critical components of the Transformer architecture, we identify tokenization strategy and multihead attention as the two major components driving brain alignment. A simple form of recurrence further improves alignment. We further demonstrate this quantitative brain alignment of our model by reproducing landmark studies in the language neuroscience field, showing that localized model units -- just like language voxels measured empirically in the human brain -- discriminate more reliably between lexical than syntactic differences, and exhibit similar response profiles under the same experimental conditions. Finally, we demonstrate the utility of our model's representations for language modeling, achieving improved sample and parameter efficiency over comparable architectures. Our model's estimates of surprisal sets a new state-of-the-art in the behavioral alignment to human reading times. Taken together, we propose a highly brain- and behaviorally-aligned model that conceptualizes the human language system as an untrained shallow feature encoder, with structural priors, combined with a trained decoder to achieve efficient and performant language processing.
Authors: Yanis Labrak, Gabriel Meseguer-Brocal, Elena V. Epure
Abstract: In recent years, generated content in music has gained significant popularity, with large language models being effectively utilized to produce human-like lyrics in various styles, themes, and linguistic structures. This technological advancement supports artists in their creative processes but also raises issues of authorship infringement, consumer satisfaction and content spamming. To address these challenges, methods for detecting generated lyrics are necessary. However, existing works have not yet focused on this specific modality or on creative text in general regarding machine-generated content detection methods and datasets. In response, we have curated the first dataset of high-quality synthetic lyrics and conducted a comprehensive quantitative evaluation of various few-shot content detection approaches, testing their generalization capabilities and complementing this with a human evaluation. Our best few-shot detector, based on LLM2Vec, surpasses stylistic and statistical methods, which are shown competitive in other domains at distinguishing human-written from machine-generated content. It also shows good generalization capabilities to new artists and models, and effectively detects post-generation paraphrasing. This study emphasizes the need for further research on creative content detection, particularly in terms of generalization and scalability with larger song catalogs. All datasets, pre-processing scripts, and code are available publicly on GitHub and Hugging Face under the Apache 2.0 license.
Authors: Qingyang Zhu, Xiang Hu, Pengyu Ji, Wei Wu, Kewei Tu
Abstract: As a cornerstone in language modeling, tokenization involves segmenting text inputs into pre-defined atomic units. Conventional statistical tokenizers often disrupt constituent boundaries within words, thereby corrupting semantic information. To address this drawback, we introduce morphological structure guidance to tokenization and propose a deep model to induce character-level structures of words. Specifically, the deep model jointly encodes internal structures and representations of words with a mechanism named $\textit{MorphOverriding}$ to ensure the indecomposability of morphemes. By training the model with self-supervised objectives, our method is capable of inducing character-level structures that align with morphological rules without annotated training data. Based on the induced structures, our algorithm tokenizes words through vocabulary matching in a top-down manner. Empirical results indicate that the proposed method effectively retains complete morphemes and outperforms widely adopted methods such as BPE and WordPiece on both morphological segmentation tasks and language modeling tasks. The code will be released later.
Authors: Fatemeh Jamshidi, Gary Pike, Amit Das, Richard Chapman
Abstract: In the domain of Music Information Retrieval (MIR), Automatic Music Transcription (AMT) emerges as a central challenge, aiming to convert audio signals into symbolic notations like musical notes or sheet music. This systematic review accentuates the pivotal role of AMT in music signal analysis, emphasizing its importance due to the intricate and overlapping spectral structure of musical harmonies. Through a thorough examination of existing machine learning techniques utilized in AMT, we explore the progress and constraints of current models and methodologies. Despite notable advancements, AMT systems have yet to match the accuracy of human experts, largely due to the complexities of musical harmonies and the need for nuanced interpretation. This review critically evaluates both fully automatic and semi-automatic AMT systems, emphasizing the importance of minimal user intervention and examining various methodologies proposed to date. By addressing the limitations of prior techniques and suggesting avenues for improvement, our objective is to steer future research towards fully automated AMT systems capable of accurately and efficiently translating intricate audio signals into precise symbolic representations. This study not only synthesizes the latest advancements but also lays out a road-map for overcoming existing challenges in AMT, providing valuable insights for researchers aiming to narrow the gap between current systems and human-level transcription accuracy.
Authors: Nadav Orzech, Yotam Nitzan, Ulysse Mizrahi, Dov Danon, Amit H. Bermano
Abstract: Virtual Try-On (VTON) is a highly active line of research, with increasing demand. It aims to replace a piece of garment in an image with one from another, while preserving person and garment characteristics as well as image fidelity. Current literature takes a supervised approach for the task, impairing generalization and imposing heavy computation. In this paper, we present a novel zero-shot training-free method for inpainting a clothing garment by reference. Our approach employs the prior of a diffusion model with no additional training, fully leveraging its native generalization capabilities. The method employs extended attention to transfer image information from reference to target images, overcoming two significant challenges. We first initially warp the reference garment over the target human using deep features, alleviating "texture sticking". We then leverage the extended attention mechanism with careful masking, eliminating leakage of reference background and unwanted influence. Through a user study, qualitative, and quantitative comparison to state-of-the-art approaches, we demonstrate superior image quality and garment preservation compared unseen clothing pieces or human figures.
Authors: Brandon Huang, Chancharik Mitra, Assaf Arbelle, Leonid Karlinsky, Trevor Darrell, Roei Herzig
Abstract: The recent success of interleaved Large Multimodal Models (LMMs) in few-shot learning suggests that in-context learning (ICL) with many examples can be promising for learning new tasks. However, this many-shot multimodal ICL setting has one crucial problem: it is fundamentally limited by the model's context length set at pretraining. The problem is especially prominent in the multimodal domain, which processes both text and images, requiring additional tokens. This motivates the need for a multimodal method to compress many shots into fewer tokens without finetuning. In this work, we enable LMMs to perform multimodal, many-shot in-context learning by leveraging Multimodal Task Vectors (MTV)--compact implicit representations of in-context examples compressed in the model's attention heads. Specifically, we first demonstrate the existence of such MTV in LMMs and then leverage these extracted MTV to enable many-shot in-context learning for various vision-and-language tasks. Our experiments suggest that MTV can scale in performance with the number of compressed shots and generalize to similar out-of-domain tasks without additional context length for inference.
Authors: Daniel Dauner, Marcel Hallgarten, Tianyu Li, Xinshuo Weng, Zhiyu Huang, Zetong Yang, Hongyang Li, Igor Gilitschenski, Boris Ivanovic, Marco Pavone, Andreas Geiger, Kashyap Chitta
Abstract: Benchmarking vision-based driving policies is challenging. On one hand, open-loop evaluation with real data is easy, but these results do not reflect closed-loop performance. On the other, closed-loop evaluation is possible in simulation, but is hard to scale due to its significant computational demands. Further, the simulators available today exhibit a large domain gap to real data. This has resulted in an inability to draw clear conclusions from the rapidly growing body of research on end-to-end autonomous driving. In this paper, we present NAVSIM, a middle ground between these evaluation paradigms, where we use large datasets in combination with a non-reactive simulator to enable large-scale real-world benchmarking. Specifically, we gather simulation-based metrics, such as progress and time to collision, by unrolling bird's eye view abstractions of the test scenes for a short simulation horizon. Our simulation is non-reactive, i.e., the evaluated policy and environment do not influence each other. As we demonstrate empirically, this decoupling allows open-loop metric computation while being better aligned with closed-loop evaluations than traditional displacement errors. NAVSIM enabled a new competition held at CVPR 2024, where 143 teams submitted 463 entries, resulting in several new insights. On a large set of challenging scenarios, we observe that simple methods with moderate compute requirements such as TransFuser can match recent large-scale end-to-end driving architectures such as UniAD. Our modular framework can potentially be extended with new datasets, data curation strategies, and metrics, and will be continually maintained to host future challenges. Our code is available at https://github.com/autonomousvision/navsim.
Authors: Adam Fisch, Tommi Jaakkola, Regina Barzilay
Abstract: Selective classification allows models to abstain from making predictions (e.g., say "I don't know") when in doubt in order to obtain better effective accuracy. While typical selective models can be effective at producing more accurate predictions on average, they may still allow for wrong predictions that have high confidence, or skip correct predictions that have low confidence. Providing calibrated uncertainty estimates alongside predictions -- probabilities that correspond to true frequencies -- can be as important as having predictions that are simply accurate on average. However, uncertainty estimates can be unreliable for certain inputs. In this paper, we develop a new approach to selective classification in which we propose a method for rejecting examples with "uncertain" uncertainties. By doing so, we aim to make predictions with {well-calibrated} uncertainty estimates over the distribution of accepted examples, a property we call selective calibration. We present a framework for learning selectively calibrated models, where a separate selector network is trained to improve the selective calibration error of a given base model. In particular, our work focuses on achieving robust calibration, where the model is intentionally designed to be tested on out-of-domain data. We achieve this through a training strategy inspired by distributionally robust optimization, in which we apply simulated input perturbations to the known, in-domain training data. We demonstrate the empirical effectiveness of our approach on multiple image classification and lung cancer risk assessment tasks.
Authors: Soumyajit Guin, Shalabh Bhatnagar
Abstract: The infinite horizon setting is widely adopted for problems of reinforcement learning (RL). These invariably result in stationary policies that are optimal. In many situations, finite horizon control problems are of interest and for such problems, the optimal policies are time-varying in general. Another setting that has become popular in recent times is of Constrained Reinforcement Learning, where the agent maximizes its rewards while it also aims to satisfy some given constraint criteria. However, this setting has only been studied in the context of infinite horizon MDPs where stationary policies are optimal. We present an algorithm for constrained RL in the Finite Horizon Setting where the horizon terminates after a fixed (finite) time. We use function approximation in our algorithm which is essential when the state and action spaces are large or continuous and use the policy gradient method to find the optimal policy. The optimal policy that we obtain depends on the stage and so is non-stationary in general. To the best of our knowledge, our paper presents the first policy gradient algorithm for the finite horizon setting with constraints. We show the convergence of our algorithm to a constrained optimal policy. We also compare and analyze the performance of our algorithm through experiments and show that our algorithm performs better than some other well known algorithms.
Authors: Jiajun Zhou, Chenxuan Xie, Shengbo Gong, Zhenyu Wen, Xiangyu Zhao, Qi Xuan, Xiaoniu Yang
Abstract: In recent years, graph representation learning has achieved remarkable success while suffering from low-quality data problems. As a mature technology to improve data quality in computer vision, data augmentation has also attracted increasing attention in graph domain. To advance research in this emerging direction, this survey provides a comprehensive review and summary of existing graph data augmentation (GDAug) techniques. Specifically, this survey first provides an overview of various feasible taxonomies and categorizes existing GDAug studies based on multi-scale graph elements. Subsequently, for each type of GDAug technique, this survey formalizes standardized technical definition, discuss the technical details, and provide schematic illustration. The survey also reviews domain-specific graph data augmentation techniques, including those for heterogeneous graphs, temporal graphs, spatio-temporal graphs, and hypergraphs. In addition, this survey provides a summary of available evaluation metrics and design guidelines for graph data augmentation. Lastly, it outlines the applications of GDAug at both the data and model levels, discusses open issues in the field, and looks forward to future directions. The latest advances in GDAug are summarized in GitHub.
Authors: Mimoun Mohamed (QARMA, I2M), Fran\c{c}ois Malgouyres (IMT), Valentin Emiya (QARMA), Caroline Chaux (IPAL)
Abstract: The {\it straight-through estimator} (STE) is commonly used to optimize quantized neural networks, yet its contexts of effective performance are still unclear despite empirical successes.To make a step forward in this comprehension, we apply STE to a well-understood problem: {\it sparse support recovery}. We introduce the {\it Support Exploration Algorithm} (SEA), a novel algorithm promoting sparsity, and we analyze its performance in support recovery (a.k.a. model selection) problems. SEA explores more supports than the state-of-the-art, leading to superior performance in experiments, especially when the columns of $A$ are strongly coherent.The theoretical analysis considers recovery guarantees when the linear measurements matrix $A$ satisfies the {\it Restricted Isometry Property} (RIP).The sufficient conditions of recovery are comparable but more stringent than those of the state-of-the-art in sparse support recovery. Their significance lies mainly in their applicability to an instance of the STE.
Authors: Ziliang Xiong, Arvi Jonnarth, Abdelrahman Eldesokey, Joakim Johnander, Bastian Wandt, Per-Erik Forssen
Abstract: Computer vision systems that are deployed in safety-critical applications need to quantify their output uncertainty. We study regression from images to parameter values and here it is common to detect uncertainty by predicting probability distributions. In this context, we investigate the regression-by-classification paradigm which can represent multimodal distributions, without a prior assumption on the number of modes. Through experiments on a specifically designed synthetic dataset, we demonstrate that traditional loss functions lead to poor probability distribution estimates and severe overconfidence, in the absence of full ground truth distributions. In order to alleviate these issues, we propose hinge-Wasserstein -- a simple improvement of the Wasserstein loss that reduces the penalty for weak secondary modes during training. This enables prediction of complex distributions with multiple modes, and allows training on datasets where full ground truth distributions are not available. In extensive experiments, we show that the proposed loss leads to substantially better uncertainty estimation on two challenging computer vision tasks: horizon line detection and stereo disparity estimation.
Authors: Federico Berto, Chuanbo Hua, Junyoung Park, Laurin Luttmann, Yining Ma, Fanchen Bu, Jiarui Wang, Haoran Ye, Minsu Kim, Sanghyeok Choi, Nayeli Gast Zepeda, Andr\'e Hottung, Jianan Zhou, Jieyi Bi, Yu Hu, Fei Liu, Hyeonah Kim, Jiwoo Son, Haeyeon Kim, Davide Angioni, Wouter Kool, Zhiguang Cao, Qingfu Zhang, Joungho Kim, Jie Zhang, Kijung Shin, Cathy Wu, Sungsoo Ahn, Guojie Song, Changhyun Kwon, Kevin Tierney, Lin Xie, Jinkyoo Park
Abstract: Deep reinforcement learning (RL) has recently shown significant benefits in solving combinatorial optimization (CO) problems, reducing reliance on domain expertise, and improving computational efficiency. However, the field lacks a unified benchmark for easy development and standardized comparison of algorithms across diverse CO problems. To fill this gap, we introduce RL4CO, a unified and extensive benchmark with in-depth library coverage of 23 state-of-the-art methods and more than 20 CO problems. Built on efficient software libraries and best practices in implementation, RL4CO features modularized implementation and flexible configuration of diverse RL algorithms, neural network architectures, inference techniques, and environments. RL4CO allows researchers to seamlessly navigate existing successes and develop their unique designs, facilitating the entire research process by decoupling science from heavy engineering. We also provide extensive benchmark studies to inspire new insights and future work. RL4CO has attracted numerous researchers in the community and is open-sourced at https://github.com/ai4co/rl4co.
Authors: Marin Vlastelica, Jin Cheng, Georg Martius, Pavel Kolev
Abstract: There has been significant recent progress in the area of unsupervised skill discovery, utilizing various information-theoretic objectives as measures of diversity. Despite these advances, challenges remain: current methods require significant online interaction, fail to leverage vast amounts of available task-agnostic data and typically lack a quantitative measure of skill utility. We address these challenges by proposing a principled offline algorithm for unsupervised skill discovery that, in addition to maximizing diversity, ensures that each learned skill imitates state-only expert demonstrations to a certain degree. Our main analytical contribution is to connect Fenchel duality, reinforcement learning, and unsupervised skill discovery to maximize a mutual information objective subject to KL-divergence state occupancy constraints. Furthermore, we demonstrate the effectiveness of our method on the standard offline benchmark D4RL and on a custom offline dataset collected from a 12-DoF quadruped robot for which the policies trained in simulation transfer well to the real robotic system.
Authors: A. Martina Neuman, Jason J. Bramburger
Abstract: Graph neural networks (GNNs) have become powerful tools for processing graph-based information in various domains. A desirable property of GNNs is transferability, where a trained network can swap in information from a different graph without retraining and retain its accuracy. A recent method of capturing transferability of GNNs is through the use of graphons, which are symmetric, measurable functions representing the limit of large dense graphs. In this work, we contribute to the application of graphons to GNNs by presenting an explicit two-layer graphon neural network (WNN) architecture. We prove its ability to approximate bandlimited graphon signals within a specified error tolerance using a minimal number of network weights. We then leverage this result, to establish the transferability of an explicit two-layer GNN over all sufficiently large graphs in a convergent sequence. Our work addresses transferability between both deterministic weighted graphs and simple random graphs and overcomes issues related to the curse of dimensionality that arise in other GNN results. The proposed WNN and GNN architectures offer practical solutions for handling graph data of varying sizes while maintaining performance guarantees without extensive retraining.
Authors: Xiaowei Gao, Xinke Jiang, Dingyi Zhuang, Huanfa Chen, Shenhao Wang, Stephen Law, James Haworth
Abstract: Traffic accidents present substantial challenges to human safety and socioeconomic development in urban areas. Developing a reliable and responsible traffic accident prediction model is crucial to addressing growing public safety concerns and enhancing the safety of urban mobility systems. Traditional methods face limitations at fine spatiotemporal scales due to the sporadic nature of highrisk accidents and the predominance of nonaccident characteristics. Furthermore, while most current models show promising occurrence prediction, they overlook the uncertainties arising from the inherent nature of accidents, and then fail to adequately map the hierarchical ranking of accident risk values for more precise insights. To address these issues, we introduce the Spatiotemporal ZeroInflated Tweedie Graph Neural Network ,STZITDGNN, the first uncertainty-aware probabilistic graph deep learning model in roadlevel traffic accident prediction for multi-steps. This model integrates the interpretability of the statistical Tweedie family model and the expressive power of graph neural networks. Its decoder innovatively employs a compound Tweedie model, a Poisson distribution to model the frequency of accident occurrences and a Gamma distribution to assess injury severity, supplemented by a zeroinflated component to effectively identify exessive non-incident instances. Empirical tests using realworld traffic data from London, UK, demonstrate that the STZITDGNN surpasses other baseline models across multiple benchmarks and metrics, including accident risk value prediction, uncertainty minimisation, nonaccident road identification and accident occurrence accuracy. Our study demonstrates that STZTIDGNN can effectively inform targeted road monitoring, thereby improving urban road safety strategies.
Authors: Trevor McInroe, Adam Jelley, Stefano V. Albrecht, Amos Storkey
Abstract: Offline pretraining with a static dataset followed by online fine-tuning (offline-to-online, or OtO) is a paradigm well matched to a real-world RL deployment process. In this scenario, we aim to find the best-performing policy within a limited budget of online interactions. Previous work in the OtO setting has focused on correcting for bias introduced by the policy-constraint mechanisms of offline RL algorithms. Such constraints keep the learned policy close to the behavior policy that collected the dataset, but we show this can unnecessarily limit policy performance if the behavior policy is far from optimal. Instead, we forgo constraints and frame OtO RL as an exploration problem that aims to maximize the benefit of online data-collection. We first study the major online RL exploration methods based on intrinsic rewards and UCB in the OtO setting, showing that intrinsic rewards add training instability through reward-function modification, and UCB methods are myopic and it is unclear which learned-component's ensemble to use for action selection. We then introduce an algorithm for planning to go out-of-distribution (PTGOOD) that avoids these issues. PTGOOD uses a non-myopic planning procedure that targets exploration in relatively high-reward regions of the state-action space unlikely to be visited by the behavior policy. By leveraging concepts from the Conditional Entropy Bottleneck, PTGOOD encourages data collected online to provide new information relevant to improving the final deployment policy without altering rewards. We show empirically in several continuous control tasks that PTGOOD significantly improves agent returns during online fine-tuning and avoids the suboptimal policy convergence that many of our baselines exhibit in several environments.
Authors: Rishi Shah, Krishnanshu Jain, Sahil Manchanda, Sourav Medya, Sayan Ranu
Abstract: Graph partitioning aims to divide a graph into disjoint subsets while optimizing a specific partitioning objective. The majority of formulations related to graph partitioning exhibit NP-hardness due to their combinatorial nature. Conventional methods, like approximation algorithms or heuristics, are designed for distinct partitioning objectives and fail to achieve generalization across other important partitioning objectives. Recently machine learning-based methods have been developed that learn directly from data. Further, these methods have a distinct advantage of utilizing node features that carry additional information. However, these methods assume differentiability of target partitioning objective functions and cannot generalize for an unknown number of partitions, i.e., they assume the number of partitions is provided in advance. In this study, we develop NeuroCUT with two key innovations over previous methodologies. First, by leveraging a reinforcement learning-based framework over node representations derived from a graph neural network and positional features, NeuroCUT can accommodate any optimization objective, even those with non-differentiable functions. Second, we decouple the parameter space and the partition count making NeuroCUT inductive to any unseen number of partition, which is provided at query time. Through empirical evaluation, we demonstrate that NeuroCUT excels in identifying high-quality partitions, showcases strong generalization across a wide spectrum of partitioning objectives, and exhibits strong generalization to unseen partition count.
Authors: Elan Rosenfeld, Nir Rosenfeld
Abstract: The goal of strategic classification is to learn decision rules which are robust to strategic input manipulation. Earlier works assume that these responses are known; while some recent works handle unknown responses, they exclusively study online settings with repeated model deployments. But there are many domains$\unicode{x2014}$particularly in public policy, a common motivating use case$\unicode{x2014}$where multiple deployments are infeasible, or where even one bad round is unacceptable. To address this gap, we initiate the formal study of one-shot strategic classification under unknown responses, which requires committing to a single classifier once. Focusing on uncertainty in the users' cost function, we begin by proving that for a broad class of costs, even a small mis-estimation of the true cost can entail trivial accuracy in the worst case. In light of this, we frame the task as a minimax problem, aiming to minimize worst-case risk over an uncertainty set of costs. We design efficient algorithms for both the full-batch and stochastic settings, which we prove converge (offline) to the minimax solution at the rate of $\tilde{\mathcal{O}}(T^{-\frac{1}{2}})$. Our analysis reveals important structure stemming from strategic responses, particularly the value of dual norm regularization with respect to the cost function.
Authors: Timm Hess, Tinne Tuytelaars, Gido M. van de Ven
Abstract: Recent years have seen considerable progress in the continual training of deep neural networks, predominantly thanks to approaches that add replay or regularization terms to the loss function to approximate the joint loss over all tasks so far. However, we show that even with a perfect approximation to the joint loss, these approaches still suffer from temporary but substantial forgetting when starting to train on a new task. Motivated by this 'stability gap', we propose that continual learning strategies should focus not only on the optimization objective, but also on the way this objective is optimized. While there is some continual learning work that alters the optimization trajectory (e.g., using gradient projection techniques), this line of research is positioned as alternative to improving the optimization objective, while we argue it should be complementary. In search of empirical support for our proposition, we perform a series of pre-registered experiments combining replay-approximated joint objectives with gradient projection-based optimization routines. However, this first experimental attempt fails to show clear and consistent benefits. Nevertheless, our conceptual arguments, as well as some of our empirical results, demonstrate the distinctive importance of the optimization trajectory in continual learning, thereby opening up a new direction for continual learning research.
Authors: Alexej Brauer
Abstract: Currently, there is a lot of research in the field of neural networks for non-life insurance pricing. The usual goal is to improve the predictive power via neural networks while building upon the generalized linear model, which is the current industry standard. Our paper contributes to this current journey via novel methods to enhance actuarial non-life models with transformer models for tabular data. We build here upon the foundation laid out by the combined actuarial neural network as well as the localGLMnet and enhance those models via the feature tokenizer transformer. The manuscript demonstrates the performance of the proposed methods on a real-world claim frequency dataset and compares them with several benchmark models such as generalized linear models, feed-forward neural networks, combined actuarial neural networks, LocalGLMnet, and pure feature tokenizer transformer. The paper shows that the new methods can achieve better results than the benchmark models while preserving certain generalized linear model advantages. The paper also discusses the practical implications and challenges of applying transformer models in actuarial settings.
Authors: Katharina Limbeck, Rayna Andreeva, Rik Sarkar, Bastian Rieck
Abstract: The magnitude of a metric space is a novel invariant that provides a measure of the 'effective size' of a space across multiple scales, while also capturing numerous geometrical properties, such as curvature, density, or entropy. We develop a family of magnitude-based measures of the intrinsic diversity of latent representations, formalising a novel notion of dissimilarity between magnitude functions of finite metric spaces. Our measures are provably stable under perturbations of the data, can be efficiently calculated, and enable a rigorous multi-scale characterisation and comparison of latent representations. We show their utility and superior performance across different domains and tasks, including (i) the automated estimation of diversity, (ii) the detection of mode collapse, and (iii) the evaluation of generative models for text, image, and graph data.
Authors: Xiran Qu, Xuequn Shang, Yupei Zhang
Abstract: This paper studies the problem of CPRP, concept prerequisite relation prediction, which is a fundamental task in using AI for education. CPRP is usually formulated into a link-prediction task on a relationship graph of concepts and solved by training the graph neural network (GNN) model. However, current directed GNNs fail to manage graph isomorphism which refers to the invariance of non-isomorphic graphs, reducing the expressivity of resulting representations. We present a permutation-equivariant directed GNN model by introducing the Weisfeiler-Lehman test into directed GNN learning. Our method is then used for CPRP and evaluated on three public datasets. The experimental results show that our model delivers better prediction performance than the state-of-the-art methods.
Authors: Erik Schultheis, Wojciech Kot{\l}owski, Marek Wydmuch, Rohit Babbar, Strom Borman, Krzysztof Dembczy\'nski
Abstract: We consider the optimization of complex performance metrics in multi-label classification under the population utility framework. We mainly focus on metrics linearly decomposable into a sum of binary classification utilities applied separately to each label with an additional requirement of exactly $k$ labels predicted for each instance. These "macro-at-$k$" metrics possess desired properties for extreme classification problems with long tail labels. Unfortunately, the at-$k$ constraint couples the otherwise independent binary classification tasks, leading to a much more challenging optimization problem than standard macro-averages. We provide a statistical framework to study this problem, prove the existence and the form of the optimal classifier, and propose a statistically consistent and practical learning algorithm based on the Frank-Wolfe method. Interestingly, our main results concern even more general metrics being non-linear functions of label-wise confusion matrices. Empirical results provide evidence for the competitive performance of the proposed approach.
Authors: Toni J. B. Liu, Nicolas Boull\'e, Rapha\"el Sarfati, Christopher J. Earls
Abstract: Pretrained large language models (LLMs) are surprisingly effective at performing zero-shot tasks, including time-series forecasting. However, understanding the mechanisms behind such capabilities remains highly challenging due to the complexity of the models. We study LLMs' ability to extrapolate the behavior of dynamical systems whose evolution is governed by principles of physical interest. Our results show that LLaMA 2, a language model trained primarily on texts, achieves accurate predictions of dynamical system time series without fine-tuning or prompt engineering. Moreover, the accuracy of the learned physical rules increases with the length of the input context window, revealing an in-context version of neural scaling law. Along the way, we present a flexible and efficient algorithm for extracting probability density functions of multi-digit numbers directly from LLMs.
Authors: Xinyi Wang, Alfonso Amayuelas, Kexun Zhang, Liangming Pan, Wenhu Chen, William Yang Wang
Abstract: Pre-trained language models (LMs) are able to perform complex reasoning without explicit fine-tuning. To understand how pre-training with a next-token prediction objective contributes to the emergence of such reasoning capability, we propose that we can view an LM as deriving new conclusions by aggregating indirect reasoning paths seen at pre-training time. We found this perspective effective in two important cases of reasoning: logic reasoning with knowledge graphs (KGs) and chain-of-thought (CoT) reasoning. More specifically, we formalize the reasoning paths as random walk paths on the knowledge/reasoning graphs. Analyses of learned LM distributions suggest that a weighted sum of relevant random walk path probabilities is a reasonable way to explain how LMs reason. Experiments and analysis on multiple KG and CoT datasets reveal the effect of training on random walk paths and suggest that augmenting unlabeled random walk reasoning paths can improve real-world multi-step reasoning performance. code: https://github.com/WANGXinyiLinda/LM_random_walk
Authors: Baohao Liao, Christian Herold, Shahram Khadivi, Christof Monz
Abstract: Memory-efficient finetuning of large language models (LLMs) has recently attracted huge attention with the increasing size of LLMs, primarily due to the constraints posed by GPU memory limitations and the effectiveness of these methods compared to full finetuning. Despite the advancements, current strategies for memory-efficient finetuning, such as QLoRA, exhibit inconsistent performance across diverse bit-width quantizations and multifaceted tasks. This inconsistency largely stems from the detrimental impact of the quantization process on preserved knowledge, leading to catastrophic forgetting and undermining the utilization of pretrained models for finetuning purposes. In this work, we introduce a novel quantization framework, ApiQ, designed to restore the lost information from quantization by concurrently initializing the LoRA components and quantizing the weights of LLMs. This approach ensures the maintenance of the original LLM's activation precision while mitigating the error propagation from shallower into deeper layers. Through comprehensive evaluations conducted on a spectrum of language tasks with various LLMs, ApiQ demonstrably minimizes activation error during quantization. Consequently, it consistently achieves superior finetuning results across various bit-widths.
Authors: Weikang Wan, Yufei Wang, Zackory Erickson, David Held
Abstract: This paper introduces DiffTOP, which utilizes Differentiable Trajectory OPtimization as the policy representation to generate actions for deep reinforcement and imitation learning. Trajectory optimization is a powerful and widely used algorithm in control, parameterized by a cost and a dynamics function. The key to our approach is to leverage the recent progress in differentiable trajectory optimization, which enables computing the gradients of the loss with respect to the parameters of trajectory optimization. As a result, the cost and dynamics functions of trajectory optimization can be learned end-to-end. DiffTOP addresses the ``objective mismatch'' issue of prior model-based RL algorithms, as the dynamics model in DiffTOP is learned to directly maximize task performance by differentiating the policy gradient loss through the trajectory optimization process. We further benchmark DiffTOP for imitation learning on standard robotic manipulation task suites with high-dimensional sensory observations and compare our method to feed-forward policy classes as well as Energy-Based Models (EBM) and Diffusion. Across 15 model-based RL tasks and 35imitation learning tasks with high-dimensional image and point cloud inputs, DiffTOP outperforms prior state-of-the-art methods in both domains.
Authors: Mohammed Muqeeth, Haokun Liu, Yufan Liu, Colin Raffel
Abstract: Recently, there has been a widespread proliferation of "expert" language models that are specialized to a specific task or domain through parameter-efficient fine-tuning. How can we recycle large collections of expert language models to improve zero-shot generalization to unseen tasks? In this work, we propose Post-Hoc Adaptive Tokenwise Gating Over an Ocean of Specialized Experts (PHATGOOSE), which learns to route among specialized modules that were produced through parameter-efficient fine-tuning. Unlike past methods that learn to route among specialized models, PHATGOOSE explores the possibility that zero-shot generalization will be improved if different experts can be adaptively chosen for each token and at each layer in the model. Crucially, our method is post-hoc - it does not require simultaneous access to the datasets used to create the specialized models and only requires a modest amount of additional compute after each expert model is trained. In experiments covering a range of specialized model collections and zero-shot generalization benchmarks, we find that PHATGOOSE outperforms past methods for post-hoc routing and, in some cases, outperforms explicit multitask training (which requires simultaneous data access). To better understand the routing strategy learned by PHATGOOSE, we perform qualitative experiments to validate that PHATGOOSE's performance stems from its ability to make adaptive per-token and per-module expert choices. We release all of our code to support future work on improving zero-shot generalization by recycling specialized experts.
Authors: Akshay Kumar, Jarvis Haupt
Abstract: This paper examines gradient flow dynamics of two-homogeneous neural networks for small initializations, where all weights are initialized near the origin. For both square and logistic losses, it is shown that for sufficiently small initializations, the gradient flow dynamics spend sufficient time in the neighborhood of the origin to allow the weights of the neural network to approximately converge in direction to the Karush-Kuhn-Tucker (KKT) points of a neural correlation function that quantifies the correlation between the output of the neural network and corresponding labels in the training data set. For square loss, it has been observed that neural networks undergo saddle-to-saddle dynamics when initialized close to the origin. Motivated by this, this paper also shows a similar directional convergence among weights of small magnitude in the neighborhood of certain saddle points.
Authors: Yeonhong Park, Jake Hyun, SangLyul Cho, Bonggeun Sim, Jae W. Lee
Abstract: Recently, considerable efforts have been directed towards compressing Large Language Models (LLMs), which showcase groundbreaking capabilities across diverse applications but entail significant deployment costs due to their large sizes. Meanwhile, much less attention has been given to mitigating the costs associated with deploying multiple LLMs of varying sizes despite its practical significance. Thus, this paper introduces \emph{any-precision LLM}, extending the concept of any-precision DNN to LLMs. Addressing challenges in any-precision LLM, we propose a lightweight method for any-precision quantization of LLMs, leveraging a post-training quantization framework, and develop a specialized software engine for its efficient serving. As a result, our solution significantly reduces the high costs of deploying multiple, different-sized LLMs by overlaying LLMs quantized to varying bit-widths, such as 3, 4, ..., $n$ bits, into a memory footprint comparable to a single $n$-bit LLM. All the supported LLMs with varying bit-widths demonstrate state-of-the-art model quality and inference throughput, proving itself to be a compelling option for deployment of multiple, different-sized LLMs. Our code is open-sourced and available online.
Authors: Lunjia Hu, Arun Jambulapati, Kevin Tian, Chutong Yang
Abstract: In the recent literature on machine learning and decision making, calibration has emerged as a desirable and widely-studied statistical property of the outputs of binary prediction models. However, the algorithmic aspects of measuring model calibration have remained relatively less well-explored. Motivated by [BGHN23], which proposed a rigorous framework for measuring distances to calibration, we initiate the algorithmic study of calibration through the lens of property testing. We define the problem of calibration testing from samples where given $n$ draws from a distribution $\mathcal{D}$ on $(predictions, binary outcomes)$, our goal is to distinguish between the case where $\mathcal{D}$ is perfectly calibrated, and the case where $\mathcal{D}$ is $\varepsilon$-far from calibration. We make the simple observation that the empirical smooth calibration linear program can be reformulated as an instance of minimum-cost flow on a highly-structured graph, and design an exact dynamic programming-based solver for it which runs in time $O(n\log^2(n))$, and solves the calibration testing problem information-theoretically optimally in the same time. This improves upon state-of-the-art black-box linear program solvers requiring $\Omega(n^\omega)$ time, where $\omega > 2$ is the exponent of matrix multiplication. We also develop algorithms for tolerant variants of our testing problem improving upon black-box linear program solvers, and give sample complexity lower bounds for alternative calibration measures to the one considered in this work. Finally, we present experiments showing the testing problem we define faithfully captures standard notions of calibration, and that our algorithms scale efficiently to accommodate large sample sizes.
Authors: Jonas H\"ubotter, Bhavya Sukhija, Lenart Treven, Yarden As, Andreas Krause
Abstract: We study the question: How can we select the right data for fine-tuning to a specific task? We call this data selection problem active fine-tuning and show that it is an instance of transductive active learning, a novel generalization of classical active learning. We propose ITL, short for information-based transductive learning, an approach which samples adaptively to maximize information gained about the specified task. We are the first to show, under general regularity assumptions, that such decision rules converge uniformly to the smallest possible uncertainty obtainable from the accessible data. We apply ITL to the few-shot fine-tuning of large neural networks and show that fine-tuning with ITL learns the task with significantly fewer examples than the state-of-the-art.
Authors: Kate Donahue, Nicole Immorlica, Meena Jagadeesan, Brendan Lucier, Aleksandrs Slivkins
Abstract: When deployed in the world, a learning agent such as a recommender system or a chatbot often repeatedly interacts with another learning agent (such as a user) over time. In many such two-agent systems, each agent learns separately and the rewards of the two agents are not perfectly aligned. To better understand such cases, we examine the learning dynamics of the two-agent system and the implications for each agent's objective. We model these systems as Stackelberg games with decentralized learning and show that standard regret benchmarks (such as Stackelberg equilibrium payoffs) result in worst-case linear regret for at least one player. To better capture these systems, we construct a relaxed regret benchmark that is tolerant to small learning errors by agents. We show that standard learning algorithms fail to provide sublinear regret, and we develop algorithms to achieve near-optimal $O(T^{2/3})$ regret for both players with respect to these benchmarks. We further design relaxed environments under which faster learning ($O(\sqrt{T})$) is possible. Altogether, our results take a step towards assessing how two-agent interactions in sequential and decentralized learning environments affect the utility of both agents.
Authors: James Urquhart Allingham, Bruno Kacper Mlodozeniec, Shreyas Padhy, Javier Antor\'an, David Krueger, Richard E. Turner, Eric Nalisnick, Jos\'e Miguel Hern\'andez-Lobato
Abstract: Correctly capturing the symmetry transformations of data can lead to efficient models with strong generalization capabilities, though methods incorporating symmetries often require prior knowledge. While recent advancements have been made in learning those symmetries directly from the dataset, most of this work has focused on the discriminative setting. In this paper, we take inspiration from group theoretic ideas to construct a generative model that explicitly aims to capture the data's approximate symmetries. This results in a model that, given a prespecified broad set of possible symmetries, learns to what extent, if at all, those symmetries are actually present. Our model can be seen as a generative process for data augmentation. We provide a simple algorithm for learning our generative model and empirically demonstrate its ability to capture symmetries under affine and color transformations, in an interpretable way. Combining our symmetry model with standard generative models results in higher marginal test-log-likelihoods and improved data efficiency.
Authors: Dennis Grinwald, Philipp Wiesner, Shinichi Nakajima
Abstract: Statistical heterogeneity in federated learning poses two major challenges: slow global training due to conflicting gradient signals, and the need of personalization for local distributions. In this work, we tackle both challenges by leveraging recent advances in \emph{linear mode connectivity} -- identifying a linearly connected low-loss region in the weight space of neural networks, which we call solution simplex. We propose federated learning over connected modes (\textsc{Floco}), where clients are assigned local subregions in this simplex based on their gradient signals, and together learn the shared global solution simplex. This allows personalization of the client models to fit their local distributions within the degrees of freedom in the solution simplex and homogenizes the update signals for the global simplex training. Our experiments show that \textsc{Floco} accelerates the global training process, and significantly improves the local accuracy with minimal computational overhead.
Authors: Mohammad Ali Labbaf-Khaniki, Mohammad Manthouri
Abstract: Fault detection and diagnosis (FDD) is a crucial task for ensuring the safety and efficiency of industrial processes. We propose a novel FDD methodology for the Tennessee Eastman Process (TEP), a widely used benchmark for chemical process control. The model employs two separate Transformer branches, enabling independent processing of input data and potential extraction of diverse information. A novel attention mechanism, Gated Dynamic Learnable Attention (GDLAttention), is introduced which integrates a gating mechanism and dynamic learning capabilities. The gating mechanism modulates the attention weights, allowing the model to focus on the most relevant parts of the input. The dynamic learning approach adapts the attention strategy during training, potentially leading to improved performance. The attention mechanism uses a bilinear similarity function, providing greater flexibility in capturing complex relationships between query and key vectors. In order to assess the effectiveness of our approach, we tested it against 21 and 18 distinct fault scenarios in TEP, and compared its performance with several established FDD techniques. The outcomes indicate that the method outperforms others in terms of accuracy, false alarm rate, and misclassification rate. This underscores the robustness and efficacy of the approach for FDD in intricate industrial processes.
Authors: Bhrij Patel, Wesley A. Suttle, Alec Koppel, Vaneet Aggarwal, Brian M. Sadler, Amrit Singh Bedi, Dinesh Manocha
Abstract: In the context of average-reward reinforcement learning, the requirement for oracle knowledge of the mixing time, a measure of the duration a Markov chain under a fixed policy needs to achieve its stationary distribution, poses a significant challenge for the global convergence of policy gradient methods. This requirement is particularly problematic due to the difficulty and expense of estimating mixing time in environments with large state spaces, leading to the necessity of impractically long trajectories for effective gradient estimation in practical applications. To address this limitation, we consider the Multi-level Actor-Critic (MAC) framework, which incorporates a Multi-level Monte-Carlo (MLMC) gradient estimator. With our approach, we effectively alleviate the dependency on mixing time knowledge, a first for average-reward MDPs global convergence. Furthermore, our approach exhibits the tightest available dependence of $\mathcal{O}\left( \sqrt{\tau_{mix}} \right)$known from prior work. With a 2D grid world goal-reaching navigation experiment, we demonstrate that MAC outperforms the existing state-of-the-art policy gradient-based method for average reward settings.
Authors: Daniele Meli
Abstract: Online unsupervised detection of anomalies is crucial to guarantee the correct operation of cyber-physical systems and the safety of humans interacting with them. State-of-the-art approaches based on deep learning via neural networks achieve outstanding performance at anomaly recognition, evaluating the discrepancy between a normal model of the system (with no anomalies) and the real-time stream of sensor time series. However, large training data and time are typically required, and explainability is still a challenge to identify the root of the anomaly and implement predictive maintainance. In this paper, we use causal discovery to learn a normal causal graph of the system, and we evaluate the persistency of causal links during real-time acquisition of sensor data to promptly detect anomalies. On two benchmark anomaly detection datasets, we show that our method has higher training efficiency, outperforms the accuracy of state-of-the-art neural architectures and correctly identifies the sources of $>10$ different anomalies. The code for experimental replication is at http://tinyurl.com/case24causal.
Authors: Mhairi Dunion, Stefano V. Albrecht
Abstract: The performance of image-based Reinforcement Learning (RL) agents can vary depending on the position of the camera used to capture the images. Training on multiple cameras simultaneously, including a first-person egocentric camera, can leverage information from different camera perspectives to improve the performance of RL. However, hardware constraints may limit the availability of multiple cameras in real-world deployment. Additionally, cameras may become damaged in the real-world preventing access to all cameras that were used during training. To overcome these hardware constraints, we propose Multi-View Disentanglement (MVD), which uses multiple cameras to learn a policy that is robust to a reduction in the number of cameras to generalise to any single camera from the training set. Our approach is a self-supervised auxiliary task for RL that learns a disentangled representation from multiple cameras, with a shared representation that is aligned across all cameras to allow generalisation to a single camera, and a private representation that is camera-specific. We show experimentally that an RL agent trained on a single third-person camera is unable to learn an optimal policy in many control tasks; but, our approach, benefiting from multiple cameras during training, is able to solve the task using only the same single third-person camera.
Authors: Rhea Sanjay Sukthanker, Arber Zela, Benedikt Staffler, Aaron Klein, Lennart Purucker, Joerg K. H. Franke, Frank Hutter
Abstract: The increasing size of language models necessitates a thorough analysis across multiple dimensions to assess trade-offs among crucial hardware metrics such as latency, energy consumption, GPU memory usage, and performance. Identifying optimal model configurations under specific hardware constraints is becoming essential but remains challenging due to the computational load of exhaustive training and evaluation on multiple devices. To address this, we introduce HW-GPT-Bench, a hardware-aware benchmark that utilizes surrogate predictions to approximate various hardware metrics across 13 devices of architectures in the GPT-2 family, with architectures containing up to 774M parameters. Our surrogates, via calibrated predictions and reliable uncertainty estimates, faithfully model the heteroscedastic noise inherent in the energy and latency measurements. To estimate perplexity, we employ weight-sharing techniques from Neural Architecture Search (NAS), inheriting pretrained weights from the largest GPT-2 model. Finally, we demonstrate the utility of HW-GPT-Bench by simulating optimization trajectories of various multi-objective optimization algorithms in just a few seconds.
Authors: Zhiwei Zhang, Minhua Lin, Enyan Dai, Suhang Wang
Abstract: Graph Neural Networks (GNNs) have shown remarkable performance in various tasks. However, recent works reveal that GNNs are vulnerable to backdoor attacks. Generally, backdoor attack poisons the graph by attaching backdoor triggers and the target class label to a set of nodes in the training graph. A GNN trained on the poisoned graph will then be misled to predict test nodes attached with trigger to the target class. Despite their effectiveness, our empirical analysis shows that triggers generated by existing methods tend to be out-of-distribution (OOD), which significantly differ from the clean data. Hence, these injected triggers can be easily detected and pruned with widely used outlier detection methods in real-world applications. Therefore, in this paper, we study a novel problem of unnoticeable graph backdoor attacks with in-distribution (ID) triggers. To generate ID triggers, we introduce an OOD detector in conjunction with an adversarial learning strategy to generate the attributes of the triggers within distribution. To ensure a high attack success rate with ID triggers, we introduce novel modules designed to enhance trigger memorization by the victim model trained on poisoned graph. Extensive experiments on real-world datasets demonstrate the effectiveness of the proposed method in generating in distribution triggers that can by-pass various defense strategies while maintaining a high attack success rate.
Authors: Omar Erak, Omar Alhussein, Shimaa Naser, Nouf Alabbasi, De Mi, Sami Muhaidat
Abstract: This study introduces an innovative framework that employs large language models (LLMs) to automate the design and generation of curricula for reinforcement learning (RL). As mobile networks evolve towards the 6G era, managing their increasing complexity and dynamic nature poses significant challenges. Conventional RL approaches often suffer from slow convergence and poor generalization due to conflicting objectives and the large state and action spaces associated with mobile networks. To address these shortcomings, we introduce curriculum learning, a method that systematically exposes the RL agent to progressively challenging tasks, improving convergence and generalization. However, curriculum design typically requires extensive domain knowledge and manual human effort. Our framework mitigates this by utilizing the generative capabilities of LLMs to automate the curriculum design process, significantly reducing human effort while improving the RL agent's convergence and performance. We deploy our approach within a simulated mobile network environment and demonstrate improved RL convergence rates, generalization to unseen scenarios, and overall performance enhancements. As a case study, we consider autonomous coordination and user association in mobile networks. Our obtained results highlight the potential of combining LLM-based curriculum generation with RL for managing next-generation wireless networks, marking a significant step towards fully autonomous network operations.
Authors: Bernd Frauenknecht, Artur Eisele, Devdutt Subhasish, Friedrich Solowjow, Sebastian Trimpe
Abstract: Dyna-style model-based reinforcement learning (MBRL) combines model-free agents with predictive transition models through model-based rollouts. This combination raises a critical question: 'When to trust your model?'; i.e., which rollout length results in the model providing useful data? Janner et al. (2019) address this question by gradually increasing rollout lengths throughout the training. While theoretically tempting, uniform model accuracy is a fallacy that collapses at the latest when extrapolating. Instead, we propose asking the question 'Where to trust your model?'. Using inherent model uncertainty to consider local accuracy, we obtain the Model-Based Actor-Critic with Uncertainty-Aware Rollout Adaption (MACURA) algorithm. We propose an easy-to-tune rollout mechanism and demonstrate substantial improvements in data efficiency and performance compared to state-of-the-art deep MBRL methods on the MuJoCo benchmark.
Authors: Ayman Chaouki, Jesse Read, Albert Bifet
Abstract: Decision Tree Learning is a fundamental problem for Interpretable Machine Learning, yet it poses a formidable optimization challenge. Despite numerous efforts dating back to the early 1990's, practical algorithms have only recently emerged, primarily leveraging Dynamic Programming (DP) and Branch & Bound (B&B) techniques. These breakthroughs led to the development of two distinct approaches. Algorithms like DL8.5 and MurTree operate on the space of nodes (or branches), they are very fast, but do not penalise complex Decision Trees, i.e. they do not solve for sparsity. On the other hand, algorithms like OSDT and GOSDT operate on the space of Decision Trees, they solve for sparsity but at the detriment of speed. In this work, we introduce Branches, a novel algorithm that integrates the strengths of both paradigms. Leveraging DP and B&B, Branches achieves exceptional speed while also solving for sparsity. Central to its efficiency is a novel analytical bound enabling substantial pruning of the search space. Furthermore, Branches does not necessitate binary features. Theoretical analysis demonstrates that Branches has a lower complexity bound compared to state-of-the-art methods, a claim validated through extensive empirical evaluation. Our results illustrate that Branches outperforms the state of the art in terms of speed and number of iterations while consistently yielding optimal Decision Trees.
Authors: Simon Schug, Seijin Kobayashi, Yassir Akram, Jo\~ao Sacramento, Razvan Pascanu
Abstract: Transformers can under some circumstances generalize to novel problem instances whose constituent parts might have been encountered during training but whose compositions have not. What mechanisms underlie this ability for compositional generalization? By reformulating multi-head attention as a hypernetwork, we reveal that a low-dimensional latent code specifies key-query specific operations. We find empirically that this latent code is highly structured, capturing information about the subtasks performed by the network. Using the framework of attention as a hypernetwork we further propose a simple modification of multi-head linear attention that strengthens the ability for compositional generalization on a range of abstract reasoning tasks. In particular, we introduce a symbolic version of the Raven Progressive Matrices human intelligence test on which we demonstrate how scaling model size and data enables compositional generalization and gives rise to a functionally structured latent code in the transformer.
Authors: Davide Corsi, Guy Amir, Andoni Rodriguez, Cesar Sanchez, Guy Katz, Roy Fox
Abstract: In recent years, Deep Reinforcement Learning (DRL) has emerged as an effective approach to solving real-world tasks. However, despite their successes, DRL-based policies suffer from poor reliability, which limits their deployment in safety-critical domains. Various methods have been put forth to address this issue by providing formal safety guarantees. Two main approaches include shielding and verification. While shielding ensures the safe behavior of the policy by employing an external online component (i.e., a ``shield'') that overrides potentially dangerous actions, this approach has a significant computational cost as the shield must be invoked at runtime to validate every decision. On the other hand, verification is an offline process that can identify policies that are unsafe, prior to their deployment, yet, without providing alternative actions when such a policy is deemed unsafe. In this work, we present verification-guided shielding -- a novel approach that bridges the DRL reliability gap by integrating these two methods. Our approach combines both formal and probabilistic verification tools to partition the input domain into safe and unsafe regions. In addition, we employ clustering and symbolic representation procedures that compress the unsafe regions into a compact representation. This, in turn, allows to temporarily activate the shield solely in (potentially) unsafe regions, in an efficient manner. Our novel approach allows to significantly reduce runtime overhead while still preserving formal safety guarantees. We extensively evaluate our approach on two benchmarks from the robotic navigation domain, as well as provide an in-depth analysis of its scalability and completeness.
Authors: Ovanes Petrosian, Li Pengyi, He Yulong, Liu Jiarui, Sun Zhaoruikun, Fu Guofeng, Meng Liping
Abstract: Rolling bearing fault detection has developed rapidly in the field of fault diagnosis technology, and it occupies a very important position in this field. Deep learning-based bearing fault diagnosis models have achieved significant success. At the same time, with the continuous improvement of new signal processing technologies such as Fourier transform, wavelet transform and empirical mode decomposition, the fault diagnosis technology of rolling bearings has also been greatly developed, and it can be said that it has entered a new research stage. However, most of the existing methods are limited to varying degrees in the industrial field. The main ones are fast feature extraction and computational complexity. The key to this paper is to propose a lightweight bearing fault diagnosis model DKDL-Net to solve these challenges. The model is trained on the CWRU data set by decoupling knowledge distillation and low rank adaptive fine tuning. Specifically, we built and trained a teacher model based on a 6-layer neural network with 69,626 trainable parameters, and on this basis, using decoupling knowledge distillation (DKD) and Low-Rank adaptive (LoRA) fine-tuning, we trained the student sag model DKDL-Net, which has only 6838 parameters. Experiments show that DKDL-Net achieves 99.48% accuracy in computational complexity on the test set while maintaining model performance, which is 0.58% higher than the state-of-the-art (SOTA) model, and our model has lower parameters. Our code is available at Github link: https://github.com/SPBU-LiPengyi/DKDL-Net.git.
Authors: Yuchao Lin, Jacob Helwig, Shurui Gui, Shuiwang Ji
Abstract: We consider achieving equivariance in machine learning systems via frame averaging. Current frame averaging methods involve a costly sum over large frames or rely on sampling-based approaches that only yield approximate equivariance. Here, we propose Minimal Frame Averaging (MFA), a mathematical framework for constructing provably minimal frames that are exactly equivariant. The general foundations of MFA also allow us to extend frame averaging to more groups than previously considered, including the Lorentz group for describing symmetries in space-time, and the unitary group for complex-valued domains. Results demonstrate the efficiency and effectiveness of encoding symmetries via MFA across a diverse range of tasks, including $n$-body simulation, top tagging in collider physics, and relaxed energy prediction. Our code is available at https://github.com/divelab/MFA.
Authors: Tomas Kulvicius, Dajie Zhang, Luise Poustka, Sven B\"olte, Lennart Jahn, Sarah Fl\"ugge, Marc Kraft, Markus Zweckstetter, Karin Nielsen-Saines, Florentin W\"org\"otter, Peter B Marschik
Abstract: There is a recent boom in the development of AI solutions to facilitate and enhance diagnostic procedures for established clinical tools. To assess the integrity of the developing nervous system, the Prechtl general movement assessment (GMA) is recognized for its clinical value in diagnosing neurological impairments in early infancy. GMA has been increasingly augmented through machine learning approaches intending to scale-up its application, circumvent costs in the training of human assessors and further standardize classification of spontaneous motor patterns. Available deep learning tools, all of which are based on single sensor modalities, are however still considerably inferior to that of well-trained human assessors. These approaches are hardly comparable as all models are designed, trained and evaluated on proprietary/silo-data sets. With this study we propose a sensor fusion approach for assessing fidgety movements (FMs) comparing three different sensor modalities (pressure, inertial, and visual sensors). Various combinations and two sensor fusion approaches (late and early fusion) for infant movement classification were tested to evaluate whether a multi-sensor system outperforms single modality assessments. The performance of the three-sensor fusion (classification accuracy of 94.5\%) was significantly higher than that of any single modality evaluated, suggesting the sensor fusion approach is a promising avenue for automated classification of infant motor patterns. The development of a robust sensor fusion system may significantly enhance AI-based early recognition of neurofunctions, ultimately facilitating automated early detection of neurodevelopmental conditions.
Authors: Tong Xia, Abhirup Ghosh, Xinchi Qiu, Cecilia Mascolo
Abstract: Federated Learning (FL) enables model development by leveraging data distributed across numerous edge devices without transferring local data to a central server. However, existing FL methods still face challenges when dealing with scarce and label-skewed data across devices, resulting in local model overfitting and drift, consequently hindering the performance of the global model. In response to these challenges, we propose a pioneering framework called FLea, incorporating the following key components: i) A global feature buffer that stores activation-target pairs shared from multiple clients to support local training. This design mitigates local model drift caused by the absence of certain classes; ii) A feature augmentation approach based on local and global activation mix-ups for local training. This strategy enlarges the training samples, thereby reducing the risk of local overfitting; iii) An obfuscation method to minimize the correlation between intermediate activations and the source data, enhancing the privacy of shared features. To verify the superiority of FLea, we conduct extensive experiments using a wide range of data modalities, simulating different levels of local data scarcity and label skew. The results demonstrate that FLea consistently outperforms state-of-the-art FL counterparts (among 13 of the experimented 18 settings, the improvement is over 5% while concurrently mitigating the privacy vulnerabilities associated with shared features. Code is available at https://github.com/XTxiatong/FLea.git.
Authors: Fin Amin, Nirjhor Rouf, Tse-Han Pan, Md Kamal Ibn Shafi, Paul D. Franzon
Abstract: In this paper, we introduce Dreamweaver, which belongs to a new class of auto-regressive decision-making models known as large reasoning models (LRMs). Dreamweaver is designed to improve 3D floorplanning in electronic design automation (EDA) via an architecture that melds advancements in sequence-to-sequence reinforcement learning algorithms. A significant advantage of our approach is its ability to effectively reason over large discrete action spaces, which is essential for handling the numerous potential positions for various functional blocks in floorplanning. Additionally, Dreamweaver demonstrates strong performance even when trained on entirely random trajectories, showcasing its capacity to leverage sub-optimal or non-expert trajectories to enhance its results. This innovative approach contributes to streamlining the integrated circuit (IC) design flow and reducing the high computational costs typically associated with floorplanning. We evaluate its performance against a current state-of-the-art method, highlighting notable improvements.
Authors: Shuang Wu, Arash A. Amini
Abstract: We consider an online decision-making problem with a reward function defined over graph-structured data. We formally formulate the problem as an instance of graph action bandit. We then propose \texttt{GNN-TS}, a Graph Neural Network (GNN) powered Thompson Sampling (TS) algorithm which employs a GNN approximator for estimating the mean reward function and the graph neural tangent features for uncertainty estimation. We prove that, under certain boundness assumptions on the reward function, GNN-TS achieves a state-of-the-art regret bound which is (1) sub-linear of order $\tilde{\mathcal{O}}((\tilde{d} T)^{1/2})$ in the number of interaction rounds, $T$, and a notion of effective dimension $\tilde{d}$, and (2) independent of the number of graph nodes. Empirical results validate that our proposed \texttt{GNN-TS} exhibits competitive performance and scales well on graph action bandit problems.
Authors: Vaneet Aggarwal, Washim Uddin Mondal, Qinbo Bai
Abstract: Reinforcement Learning (RL) serves as a versatile framework for sequential decision-making, finding applications across diverse domains such as robotics, autonomous driving, recommendation systems, supply chain optimization, biology, mechanics, and finance. The primary objective in these applications is to maximize the average reward. Real-world scenarios often necessitate adherence to specific constraints during the learning process. This monograph focuses on the exploration of various model-based and model-free approaches for Constrained RL within the context of average reward Markov Decision Processes (MDPs). The investigation commences with an examination of model-based strategies, delving into two foundational methods - optimism in the face of uncertainty and posterior sampling. Subsequently, the discussion transitions to parametrized model-free approaches, where the primal-dual policy gradient-based algorithm is explored as a solution for constrained MDPs. The monograph provides regret guarantees and analyzes constraint violation for each of the discussed setups. For the above exploration, we assume the underlying MDP to be ergodic. Further, this monograph extends its discussion to encompass results tailored for weakly communicating MDPs, thereby broadening the scope of its findings and their relevance to a wider range of practical scenarios.
Authors: Jason Gross, Rajashree Agrawal, Thomas Kwa, Euan Ong, Chun Hei Yip, Alex Gibson, Soufiane Noubir, Lawrence Chan
Abstract: In this work, we propose using mechanistic interpretability -- techniques for reverse engineering model weights into human-interpretable algorithms -- to derive and compactly prove formal guarantees on model performance. We prototype this approach by formally proving lower bounds on the accuracy of 151 small transformers trained on a Max-of-$K$ task. We create 102 different computer-assisted proof strategies and assess their length and tightness of bound on each of our models. Using quantitative metrics, we find that shorter proofs seem to require and provide more mechanistic understanding. Moreover, we find that more faithful mechanistic understanding leads to tighter performance bounds. We confirm these connections by qualitatively examining a subset of our proofs. Finally, we identify compounding structureless noise as a key challenge for using mechanistic interpretability to generate compact proofs on model performance.
Authors: Jeffrey Li, Alex Fang, Georgios Smyrnis, Maor Ivgi, Matt Jordan, Samir Gadre, Hritik Bansal, Etash Guha, Sedrick Keh, Kushal Arora, Saurabh Garg, Rui Xin, Niklas Muennighoff, Reinhard Heckel, Jean Mercat, Mayee Chen, Suchin Gururangan, Mitchell Wortsman, Alon Albalak, Yonatan Bitton, Marianna Nezhurina, Amro Abbas, Cheng-Yu Hsieh, Dhruba Ghosh, Josh Gardner, Maciej Kilian, Hanlin Zhang, Rulin Shao, Sarah Pratt, Sunny Sanyal, Gabriel Ilharco, Giannis Daras, Kalyani Marathe, Aaron Gokaslan, Jieyu Zhang, Khyathi Chandu, Thao Nguyen, Igor Vasiljevic, Sham Kakade, Shuran Song, Sujay Sanghavi, Fartash Faghri, Sewoong Oh, Luke Zettlemoyer, Kyle Lo, Alaaeldin El-Nouby, Hadi Pouransari, Alexander Toshev, Stephanie Wang, Dirk Groeneveld, Luca Soldaini, Pang Wei Koh, Jenia Jitsev, Thomas Kollar, Alexandros G. Dimakis, Yair Carmon, Achal Dave, Ludwig Schmidt, Vaishaal Shankar
Abstract: We introduce DataComp for Language Models (DCLM), a testbed for controlled dataset experiments with the goal of improving language models. As part of DCLM, we provide a standardized corpus of 240T tokens extracted from Common Crawl, effective pretraining recipes based on the OpenLM framework, and a broad suite of 53 downstream evaluations. Participants in the DCLM benchmark can experiment with data curation strategies such as deduplication, filtering, and data mixing at model scales ranging from 412M to 7B parameters. As a baseline for DCLM, we conduct extensive experiments and find that model-based filtering is key to assembling a high-quality training set. The resulting dataset, DCLM-Baseline enables training a 7B parameter language model from scratch to 64% 5-shot accuracy on MMLU with 2.6T training tokens. Compared to MAP-Neo, the previous state-of-the-art in open-data language models, DCLM-Baseline represents a 6.6 percentage point improvement on MMLU while being trained with 40% less compute. Our baseline model is also comparable to Mistral-7B-v0.3 and Llama 3 8B on MMLU (63% & 66%), and performs similarly on an average of 53 natural language understanding tasks while being trained with 6.6x less compute than Llama 3 8B. Our results highlight the importance of dataset design for training language models and offer a starting point for further research on data curation.
Authors: Victor Priser (S2A, IDS), Pascal Bianchi (S2A, IDS), Adil Salim
Abstract: Stein Variational Gradient Descent (SVGD) is a widely used sampling algorithm that has been successfully applied in several areas of Machine Learning. SVGD operates by iteratively moving a set of interacting particles (which represent the samples) to approximate the target distribution. Despite recent studies on the complexity of SVGD and its variants, their long-time asymptotic behavior (i.e., after numerous iterations ) is still not understood in the finite number of particles regime. We study the long-time asymptotic behavior of a noisy variant of SVGD. First, we establish that the limit set of noisy SVGD for large is well-defined. We then characterize this limit set, showing that it approaches the target distribution as increases. In particular, noisy SVGD provably avoids the variance collapse observed for SVGD. Our approach involves demonstrating that the trajectories of noisy SVGD closely resemble those described by a McKean-Vlasov process.
Authors: Nirjhor Rouf, Fin Amin, Paul D. Franzon
Abstract: In this work, we present empirical results regarding the feasibility of using offline large language models (LLMs) in the context of electronic design automation (EDA). The goal is to investigate and evaluate a contemporary language model's (Llama-2-7B) ability to function as a microelectronic Q & A expert as well as its reasoning, and generation capabilities in solving microelectronic-related problems. Llama-2-7B was tested across a variety of adaptation methods, including introducing a novel low-rank knowledge distillation (LoRA-KD) scheme. Our experiments produce both qualitative and quantitative results.
Authors: Marco Fumero, Marco Pegoraro, Valentino Maiorca, Francesco Locatello, Emanuele Rodol\`a
Abstract: Neural models learn data representations that lie on low-dimensional manifolds, yet modeling the relation between these representational spaces is an ongoing challenge. By integrating spectral geometry principles into neural modeling, we show that this problem can be better addressed in the functional domain, mitigating complexity, while enhancing interpretability and performances on downstream tasks. To this end, we introduce a multi-purpose framework to the representation learning community, which allows to: (i) compare different spaces in an interpretable way and measure their intrinsic similarity; (ii) find correspondences between them, both in unsupervised and weakly supervised settings, and (iii) to effectively transfer representations between distinct spaces. We validate our framework on various applications, ranging from stitching to retrieval tasks, demonstrating that latent functional maps can serve as a swiss-army knife for representation alignment.
Authors: Chaoqi Liang, Guanglei Yang, Lifeng Qiao, Zitong Huang, Hongliang Yan, Yunchao Wei, Wangmeng Zuo
Abstract: Deep neural networks have achieved remarkable performance across various tasks when supplied with large-scale labeled data. However, the collection of labeled data can be time-consuming and labor-intensive. Semi-supervised learning (SSL), particularly through pseudo-labeling algorithms that iteratively assign pseudo-labels for self-training, offers a promising solution to mitigate the dependency of labeled data. Previous research generally applies a uniform pseudo-labeling strategy across all model layers, assuming that pseudo-labels exert uniform influence throughout. Contrasting this, our theoretical analysis and empirical experiment demonstrate feature extraction layer and linear classification layer have distinct learning behaviors in response to pseudo-labels. Based on these insights, we develop two layer-specific pseudo-label strategies, termed Grad-ReLU and Avg-Clustering. Grad-ReLU mitigates the impact of noisy pseudo-labels by removing the gradient detrimental effects of pseudo-labels in the linear classification layer. Avg-Clustering accelerates the convergence of feature extraction layer towards stable clustering centers by integrating consistent outputs. Our approach, LayerMatch, which integrates these two strategies, can avoid the severe interference of noisy pseudo-labels in the linear classification layer while accelerating the clustering capability of the feature extraction layer. Through extensive experimentation, our approach consistently demonstrates exceptional performance on standard semi-supervised learning benchmarks, achieving a significant improvement of 10.38% over baseline method and a 2.44% increase compared to state-of-the-art methods.
Authors: Shengzhe Xu, Cho-Ting Lee, Mandar Sharma, Raquib Bin Yousuf, Nikhil Muralidhar, Naren Ramakrishnan
Abstract: Large language models (LLMs) have demonstrated their prowess in generating synthetic text and images; however, their potential for generating tabular data -- arguably the most common data type in business and scientific applications -- is largely underexplored. This paper demonstrates that LLMs, used as-is, or after traditional fine-tuning, are severely inadequate as synthetic table generators. Due to the autoregressive nature of LLMs, fine-tuning with random order permutation runs counter to the importance of modeling functional dependencies, and renders LLMs unable to model conditional mixtures of distributions (key to capturing real world constraints). We showcase how LLMs can be made to overcome some of these deficiencies by making them permutation-aware.
Authors: Haocheng Ju, Haimiao Zhang, Lin Li, Xiao Li, Bin Dong
Abstract: Joint channel estimation and signal detection (JCESD) is crucial in orthogonal frequency division multiplexing (OFDM) systems, but traditional algorithms perform poorly in low signal-to-noise ratio (SNR) scenarios. Deep learning (DL) methods have been investigated, but concerns regarding computational expense and lack of validation in low-SNR settings remain. Hence, the development of a robust and low-complexity model that can deliver excellent performance across a wide range of SNRs is highly desirable. In this paper, we aim to establish a benchmark where traditional algorithms and DL methods are validated on different channel models, Doppler, and SNR settings, particularly focusing on the semi-blind setting. In particular, we propose a new DL model where the backbone network is formed by unrolling the iterative algorithm, and the hyperparameters are estimated by hypernetworks. Additionally, we adapt a lightweight DenseNet to the task of JCESD for comparison. We evaluate different methods in three aspects: generalization in terms of bit error rate (BER), robustness, and complexity. Our results indicate that DL approaches outperform traditional algorithms in the challenging low-SNR setting, while the iterative algorithm performs better in high-SNR settings. Furthermore, the iterative algorithm is more robust in the presence of carrier frequency offset, whereas DL methods excel when signals are corrupted by asymmetric Gaussian noise.
Authors: Elena Celledoni, James Jackaman, Davide Murari, Brynjulf Owren
Abstract: As supported by abundant experimental evidence, neural networks are state-of-the-art for many approximation tasks in high-dimensional spaces. Still, there is a lack of a rigorous theoretical understanding of what they can approximate, at which cost, and at which accuracy. One network architecture of practical use, especially for approximation tasks involving images, is (residual) convolutional networks. However, due to the locality of the linear operators involved in these networks, their analysis is more complicated than that of fully connected neural networks. This paper deals with approximation of time sequences where each observation is a matrix. We show that with relatively small networks, we can represent exactly a class of numerical discretizations of PDEs based on the method of lines. We constructively derive these results by exploiting the connections between discrete convolution and finite difference operators. Our network architecture is inspired by those typically adopted in the approximation of time sequences. We support our theoretical results with numerical experiments simulating the linear advection, heat, and Fisher equations.
Authors: Junyu Zhou, Shuo Huang, Han Feng, Puyu Wang, Ding-Xuan Zhou
Abstract: In this paper, we are concerned with the generalization performance of non-parametric estimation for pairwise learning. Most of the existing work requires the hypothesis space to be convex or a VC-class, and the loss to be convex. However, these restrictive assumptions limit the applicability of the results in studying many popular methods, especially kernel methods and neural networks. We significantly relax these restrictive assumptions and establish a sharp oracle inequality of the empirical minimizer with a general hypothesis space for the Lipschitz continuous pairwise losses. Our results can be used to handle a wide range of pairwise learning problems including ranking, AUC maximization, pairwise regression, and metric and similarity learning. As an application, we apply our general results to study pairwise least squares regression and derive an excess generalization bound that matches the minimax lower bound for pointwise least squares regression up to a logrithmic term. The key novelty here is to construct a structured deep ReLU neural network as an approximation of the true predictor and design the targeted hypothesis space consisting of the structured networks with controllable complexity. This successful application demonstrates that the obtained general results indeed help us to explore the generalization performance on a variety of problems that cannot be handled by existing approaches.
Authors: David A. Kreplin, Marco Roth
Abstract: Quantum neural networks (QNNs) use parameterized quantum circuits with data-dependent inputs and generate outputs through the evaluation of expectation values. Calculating these expectation values necessitates repeated circuit evaluations, thus introducing fundamental finite-sampling noise even on error-free quantum computers. We reduce this noise by introducing the variance regularization, a technique for reducing the variance of the expectation value during the quantum model training. This technique requires no additional circuit evaluations if the QNN is properly constructed. Our empirical findings demonstrate the reduced variance speeds up the training and lowers the output noise as well as decreases the number of necessary evaluations of gradient circuits. This regularization method is benchmarked on the regression of multiple functions and the potential energy surface of water. We show that in our examples, it lowers the variance by an order of magnitude on average and leads to a significantly reduced noise level of the QNN. We finally demonstrate QNN training on a real quantum device and evaluate the impact of error mitigation. Here, the optimization is feasible only due to the reduced number of necessary shots in the gradient evaluation resulting from the reduced variance.
Authors: Jun Qi, Chao-Han Huck Yang, Pin-Yu Chen, Min-Hsiu Hsieh, Hector Zenil, Jesper Tegner
Abstract: While Quantum Machine Learning (QML) is an exciting emerging area, the accuracy of the loss function still needs to be improved by the number of available qubits. Here, we reformulate the QML problem such that the approximation error (representation power) does not depend on the number of qubits. We prove that a classical-to-quantum transfer learning architecture using a Variational Quantum Circuit (VQC) improves the representation and generalization (estimation error) capabilities of the VQC model. We derive analytical bounds for the approximation and estimation error. We show that the architecture of classical-to-quantum transfer learning leverages pre-trained classical generative AI models, making it easier to find the optimal parameters for the VQC in the training stage. To validate our theoretical analysis, we perform experiments on single-dot and double-dot binary classification tasks for charge stability diagrams in semiconductor quantum dots, where the related empirical results support our theoretical findings. Our analytical and empirical results demonstrate the effectiveness of classical-to-quantum transfer learning architecture in realistic tasks. This sets the stage for accelerating QML applications beyond the current limits of available qubits.
Authors: Xinyan Hu, Meena Jagadeesan, Michael I. Jordan, Jacob Steinhardt
Abstract: In content recommender systems such as TikTok and YouTube, the platform's recommendation algorithm shapes content producer incentives. Many platforms employ online learning, which generates intertemporal incentives, since content produced today affects recommendations of future content. We study the game between producers and analyze the content created at equilibrium. We show that standard online learning algorithms, such as Hedge and EXP3, unfortunately incentivize producers to create low-quality content, where producers' effort approaches zero in the long run for typical learning rate schedules. Motivated by this negative result, we design learning algorithms that incentivize producers to invest high effort and achieve high user welfare. At a conceptual level, our work illustrates the unintended impact that a platform's learning algorithm can have on content quality and introduces algorithmic approaches to mitigating these effects.
Authors: Zhiyu Mei, Wei Fu, Jiaxuan Gao, Guangju Wang, Huanchen Zhang, Yi Wu
Abstract: The ever-growing complexity of reinforcement learning (RL) tasks demands a distributed system to efficiently generate and process a massive amount of data. However, existing open-source libraries suffer from various limitations, which impede their practical use in challenging scenarios where large-scale training is necessary. In this paper, we present a novel abstraction on the dataflows of RL training, which unifies diverse RL training applications into a general framework. Following this abstraction, we develop a scalable, efficient, and extensible distributed RL system called ReaLlyScalableRL, which allows efficient and massively parallelized training and easy development of customized algorithms. Our evaluation shows that SRL outperforms existing academic libraries, reaching at most 21x higher training throughput in a distributed setting. On learning performance, beyond performing and scaling well on common RL benchmarks with different RL algorithms, SRL can reproduce the same solution in the challenging hide-and-seek environment as reported by OpenAI with up to 5x speedup in wall-clock time. Notably, SRL is the first in the academic community to perform RL experiments at a large scale with over 15k CPU cores. SRL source code is available at: https://github.com/openpsi-project/srl .
Authors: Yun Lu, Malik Magdon-Ismail, Yu Wei, Vassilis Zikas
Abstract: Differential Privacy (DP) (and its variants) is the most common method for machine learning (ML) on privacy-sensitive data. In big data analytics, one often uses randomized sketching/aggregation algorithms to make processing high-dimensional data tractable. Intuitively, such ML algorithms should provide some inherent privacy, yet most existing DP mechanisms do not leverage or under-utilize this inherent randomness, resulting in potentially redundant noising. The motivating question of our work is: (How) can we improve the utility of DP mechanisms for randomized ML queries, by leveraging the randomness of the query itself? Towards a (positive) answer, our key contribution is (proving) what we call the NDIS theorem, a theoretical result with several practical implications. In a nutshell, NDIS is a closed-form analytic computation for the (varepsilon,delta)-indistinguishability-spectrum (IS) of two arbitrary normal distributions N1 and N2, i.e., the optimal delta (for any given varepsilon) such that N1 and N2 are (varepsilon,delta)-close according to the DP distance. The importance of the NDIS theorem lies in that (1) it yields efficient estimators for IS, and (2) it allows us to analyze DP-mechanism with normally-distributed outputs, as well as more general mechanisms by leveraging their behavior on large inputs. We apply the NDIS theorem to derive DP mechanisms for queries with normally-distributed outputs--i.e., Gaussian Random Projections (RP)--and for more general queries--i.e., Ordinary Least Squares (OLS). Compared to existing techniques, our new DP mechanisms achieve superior privacy/utility trade-offs by leveraging the randomness of the underlying algorithms. We then apply the NDIS theorem to a data-driven DP notion--in particular relative DP introduced by Lu et al. [S&P 2024]. Our method identifies the range of (varepsilon,delta) for which no additional noising is needed.
Authors: Kevin Clark, Paul Vicol, Kevin Swersky, David J Fleet
Abstract: We present Direct Reward Fine-Tuning (DRaFT), a simple and effective method for fine-tuning diffusion models to maximize differentiable reward functions, such as scores from human preference models. We first show that it is possible to backpropagate the reward function gradient through the full sampling procedure, and that doing so achieves strong performance on a variety of rewards, outperforming reinforcement learning-based approaches. We then propose more efficient variants of DRaFT: DRaFT-K, which truncates backpropagation to only the last K steps of sampling, and DRaFT-LV, which obtains lower-variance gradient estimates for the case when K=1. We show that our methods work well for a variety of reward functions and can be used to substantially improve the aesthetic quality of images generated by Stable Diffusion 1.4. Finally, we draw connections between our approach and prior work, providing a unifying perspective on the design space of gradient-based fine-tuning algorithms.
Authors: Georg W\"olflein (University of St Andrews, St Andrews, United Kingdom, Else Kr\"oner Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany), Dyke Ferber (Else Kr\"oner Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany, Department of Medical Oncology, National Center for Tumor Diseases, University Hospital Heidelberg, Heidelberg, Germany), Asier R. Meneghetti (Else Kr\"oner Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany), Omar S. M. El Nahhas (Else Kr\"oner Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany), Daniel Truhn (University Hospital Aachen, Germany), Zunamys I. Carrero (Else Kr\"oner Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany), David J. Harrison (University of St Andrews, St Andrews, United Kingdom), Ognjen Arandjelovi\'c (University of St Andrews, St Andrews, United Kingdom), Jakob Nikolas Kather (Else Kr\"oner Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, TUD Dresden University of Technology, Dresden, Germany, Department of Medical Oncology, National Center for Tumor Diseases, University Hospital Heidelberg, Heidelberg, Germany, Department of Medicine I, University Hospital Dresden, Dresden, Germany)
Abstract: Weakly supervised whole slide image classification is a key task in computational pathology, which involves predicting a slide-level label from a set of image patches constituting the slide. Constructing models to solve this task involves multiple design choices, often made without robust empirical or conclusive theoretical justification. To address this, we conduct a comprehensive benchmarking of feature extractors to answer three critical questions: 1) Is stain normalisation still a necessary preprocessing step? 2) Which feature extractors are best for downstream slide-level classification? 3) How does magnification affect downstream performance? Our study constitutes the most comprehensive evaluation of publicly available pathology feature extractors to date, involving more than 10,000 training runs across 14 feature extractors, 9 tasks, 5 datasets, 3 downstream architectures, 2 levels of magnification, and various preprocessing setups. Our findings challenge existing assumptions: 1) We observe empirically, and by analysing the latent space, that skipping stain normalisation and image augmentations does not degrade performance, while significantly reducing memory and computational demands. 2) We develop a novel evaluation metric to compare relative downstream performance, and show that the choice of feature extractor is the most consequential factor for downstream performance. 3) We find that lower-magnification slides are sufficient for accurate slide-level classification. Contrary to previous patch-level benchmarking studies, our approach emphasises clinical relevance by focusing on slide-level biomarker prediction tasks in a weakly supervised setting with external validation cohorts. Our findings stand to streamline digital pathology workflows by minimising preprocessing needs and informing the selection of feature extractors.
Authors: Haochen Zhang, Yuyang Dong, Chuan Xiao, Masafumi Oyamada
Abstract: This paper explores the utilization of LLMs for data preprocessing (DP), a crucial step in the data mining pipeline that transforms raw data into a clean format conducive to easy processing. Whereas the use of LLMs has sparked interest in devising universal solutions to DP, recent initiatives in this domain typically rely on GPT APIs, raising inevitable data breach concerns. Unlike these approaches, we consider instruction-tuning local LLMs (7 -- 13B models) as universal DP task solvers that operate on a local, single, and low-priced GPU, ensuring data security and enabling further customization. We select a collection of datasets across four representative DP tasks and construct instruction tuning data using data configuration, knowledge injection, and reasoning data distillation techniques tailored to DP. By tuning Mistral-7B, Llama 3-8B, and OpenOrca-Platypus2-13B, our models, namely, Jellyfish-7B/8B/13B, deliver competitiveness compared to GPT-3.5/4 models and strong generalizability to unseen tasks while barely compromising the base models' abilities in NLP tasks. Meanwhile, Jellyfish offers enhanced reasoning capabilities compared to GPT-3.5. Our models are available at: https://huggingface.co/NECOUDBFM/Jellyfish . Our instruction dataset is available at: https://huggingface.co/datasets/NECOUDBFM/Jellyfish-Instruct .
URLs: https://huggingface.co/NECOUDBFM/Jellyfish, https://huggingface.co/datasets/NECOUDBFM/Jellyfish-Instruct
Authors: Vishwak Srinivasan, Andre Wibisono, Ashia Wilson
Abstract: We propose a new method called the Metropolis-adjusted Mirror Langevin algorithm for approximate sampling from distributions whose support is a compact and convex set. This algorithm adds an accept-reject filter to the Markov chain induced by a single step of the Mirror Langevin algorithm (Zhang et al., 2020), which is a basic discretisation of the Mirror Langevin dynamics. Due to the inclusion of this filter, our method is unbiased relative to the target, while known discretisations of the Mirror Langevin dynamics including the Mirror Langevin algorithm have an asymptotic bias. For this algorithm, we also give upper bounds for the number of iterations taken to mix to a constrained distribution whose potential is relatively smooth, convex, and Lipschitz continuous with respect to a self-concordant mirror function. As a consequence of the reversibility of the Markov chain induced by the inclusion of the Metropolis-Hastings filter, we obtain an exponentially better dependence on the error tolerance for approximate constrained sampling. We also present numerical experiments that corroborate our theoretical findings.
Authors: Haoyang Zheng, Wei Deng, Christian Moya, Guang Lin
Abstract: Approximate Thompson sampling with Langevin Monte Carlo broadens its reach from Gaussian posterior sampling to encompass more general smooth posteriors. However, it still encounters scalability issues in high-dimensional problems when demanding high accuracy. To address this, we propose an approximate Thompson sampling strategy, utilizing underdamped Langevin Monte Carlo, where the latter is the go-to workhorse for simulations of high-dimensional posteriors. Based on the standard smoothness and log-concavity conditions, we study the accelerated posterior concentration and sampling using a specific potential function. This design improves the sample complexity for realizing logarithmic regrets from $\mathcal{\tilde O}(d)$ to $\mathcal{\tilde O}(\sqrt{d})$. The scalability and robustness of our algorithm are also empirically validated through synthetic experiments in high-dimensional bandit problems.
Authors: Daniela A. Wiepert, Rene L. Utianski, Joseph R. Duffy, John L. Stricker, Leland R. Barnard, David T. Jones, Hugo Botha
Abstract: Accurately extracting clinical information from speech is critical to the diagnosis and treatment of many neurological conditions. As such, there is interest in leveraging AI for automatic, objective assessments of clinical speech to facilitate diagnosis and treatment of speech disorders. We explore transfer learning using foundation models, focusing on the impact of layer selection for the downstream task of predicting pathological speech features. We find that selecting an optimal layer can greatly improve performance (~15.8% increase in balanced accuracy per feature as compared to worst layer, ~13.6% increase as compared to final layer), though the best layer varies by predicted feature and does not always generalize well to unseen data. A learned weighted sum offers comparable performance to the average best layer in-distribution (only ~1.2% lower) and had strong generalization for out-of-distribution data (only 1.5% lower than the average best layer).
Authors: Helen Qu, Sang Michael Xie
Abstract: Models trained on a labeled source domain (e.g., labeled images from wildlife camera traps) often generalize poorly when deployed on an out-of-distribution (OOD) target domain (e.g., images from new camera trap locations). In the domain adaptation setting where unlabeled target data is available, self-supervised pretraining (e.g., masked autoencoding or contrastive learning) is a promising method to mitigate this performance drop. Pretraining improves OOD error when the generic data augmentations used (e.g., masking or cropping) connect the source and target domains, which may be far apart in the input space. In this paper, we show on real-world tasks that standard fine-tuning after pretraining does not consistently improve OOD error over simply training from scratch on labeled source data. To better leverage pretraining for distribution shifts, we propose Connect Later: after pretraining with generic augmentations, fine-tune with targeted augmentations designed with knowledge of the distribution shift. Pretraining learns good representations within the source and target domains, while targeted augmentations connect the domains better during fine-tuning. Connect Later improves average OOD error over standard fine-tuning and supervised learning with targeted augmentations on 4 real-world datasets: Connect Later achieves the state-of-the-art on astronomical time-series classification (AstroClassification) by 2.5%, wildlife species identification (iWildCam-WILDS) with ResNet-50 by 0.9%, and tumor identification (Camelyon17-WILDS) with DenseNet121 by 1.1%; as well as best performance on a new dataset for astronomical time-series redshift prediction (Redshifts) by 0.03 RMSE (11% relative). Code and datasets are available at https://github.com/helenqu/connect-later.
Authors: Matteo Priorelli, Ivilin Peev Stoianov
Abstract: Determining an optimal plan to accomplish a goal is a hard problem in realistic scenarios, which often comprise dynamic and causal relationships between several entities. Although traditionally such problems have been tackled with optimal control and reinforcement learning, a recent biologically-motivated proposal casts planning and control as an inference process. Among these new approaches, one is particularly promising: active inference. This new paradigm assumes that action and perception are two complementary aspects of life whereby the role of the former is to fulfill the predictions inferred by the latter. In this study, we present an effective solution, based on active inference, to complex control tasks. The proposed architecture exploits hybrid (discrete and continuous) processing to construct a hierarchical and dynamic representation of the self and the environment, which is then used to produce a flexible plan consisting of subgoals at different temporal scales. We evaluate this deep hybrid model on a non-trivial task: reaching a moving object after having picked a moving tool. This study extends past work on planning as inference and advances an alternative direction to optimal control and reinforcement learning.
Authors: Paola Busia, Matteo Antonio Scrugli, Victor Jean-Baptiste Jung, Luca Benini, Paolo Meloni
Abstract: Wearable systems for the continuous and real-time monitoring of cardiovascular diseases are becoming widespread and valuable assets in diagnosis and therapy. A promising approach for real-time analysis of the electrocardiographic (ECG) signal and the detection of heart conditions, such as arrhythmia, is represented by the transformer machine learning model. Transformers are powerful models for the classification of time series, although efficient implementation in the wearable domain raises significant design challenges, to combine adequate accuracy and a suitable complexity. In this work, we present a tiny transformer model for the analysis of the ECG signal, requiring only 6k parameters and reaching 98.97% accuracy in the recognition of the 5 most common arrhythmia classes from the MIT-BIH Arrhythmia database, assessed considering 8-bit integer inference as required for efficient execution on low-power microcontroller-based devices. We explored an augmentation-based training approach for improving the robustness against electrode motion artifacts noise, resulting in a worst-case post-deployment performance assessment of 98.36% accuracy. Suitability for wearable monitoring solutions is finally demonstrated through efficient deployment on the parallel ultra-low-power GAP9 processor, where inference execution requires 4.28ms and 0.09mJ.
Authors: Oliver Bentham, Nathan Stringham, Ana Marasovi\'c
Abstract: Understanding the extent to which Chain-of-Thought (CoT) generations align with a large language model's (LLM) internal computations is critical for deciding whether to trust an LLM's output. As a proxy for CoT faithfulness, Lanham et al. (2023) propose a metric that measures a model's dependence on its CoT for producing an answer. Within a single family of proprietary models, they find that LLMs exhibit a scaling-then-inverse-scaling relationship between model size and their measure of faithfulness, and that a 13 billion parameter model exhibits increased faithfulness compared to models ranging from 810 million to 175 billion parameters in size. We evaluate whether these results generalize as a property of all LLMs. We replicate the experimental setup in their section focused on scaling experiments with three different families of models and, under specific conditions, successfully reproduce the scaling trends for CoT faithfulness they report. However, after normalizing the metric to account for a model's bias toward certain answer choices, unfaithfulness drops significantly for smaller less-capable models. This normalized faithfulness metric is also strongly correlated ($R^2$=0.74) with accuracy, raising doubts about its validity for evaluating faithfulness.
Authors: Zhenting Qi, Hanlin Zhang, Eric Xing, Sham Kakade, Himabindu Lakkaraju
Abstract: Retrieval-Augmented Generation (RAG) improves pre-trained models by incorporating external knowledge at test time to enable customized adaptation. We study the risk of datastore leakage in Retrieval-In-Context RAG Language Models (LMs). We show that an adversary can exploit LMs' instruction-following capabilities to easily extract text data verbatim from the datastore of RAG systems built with instruction-tuned LMs via prompt injection. The vulnerability exists for a wide range of modern LMs that span Llama2, Mistral/Mixtral, Vicuna, SOLAR, WizardLM, Qwen1.5, and Platypus2, and the exploitability exacerbates as the model size scales up. Extending our study to production RAG models GPTs, we design an attack that can cause datastore leakage with a 100% success rate on 25 randomly selected customized GPTs with at most 2 queries, and we extract text data verbatim at a rate of 41% from a book of 77,000 words and 3% from a corpus of 1,569,000 words by prompting the GPTs with only 100 queries generated by themselves.
Authors: Zachary Horvitz, Jingru Chen, Rahul Aditya, Harshvardhan Srivastava, Robert West, Zhou Yu, Kathleen McKeown
Abstract: Humor is a fundamental facet of human cognition and interaction. Yet, despite recent advances in natural language processing, humor detection remains a challenging task that is complicated by the scarcity of datasets that pair humorous texts with similar non-humorous counterparts. In our work, we investigate whether large language models (LLMs), can generate synthetic data for humor detection via editing texts. We benchmark LLMs on an existing human dataset and show that current LLMs display an impressive ability to 'unfun' jokes, as judged by humans and as measured on the downstream task of humor detection. We extend our approach to a code-mixed English-Hindi humor dataset, where we find that GPT-4's synthetic data is highly rated by bilingual annotators and provides challenging adversarial examples for humor classifiers.
Authors: Igor Carrara, Bruno Aristimunha, Marie-Constance Corsi, Raphael Y. de Camargo, Sylvain Chevallier, Th\'eodore Papadopoulo
Abstract: The integration of Deep Learning (DL) algorithms on brain signal analysis is still in its nascent stages compared to their success in fields like Computer Vision, especially in Brain-Computer Interface (BCI), where the brain activity is decoded to control external devices without requiring muscle control. Electroencephalography (EEG) is a widely adopted choice for designing BCI systems due to its non-invasive and cost-effective nature and excellent temporal resolution. Still, it comes at the expense of limited training data, poor signal-to-noise, and a large variability across and within-subject recordings. Finally, setting up a BCI system with many electrodes takes a long time, hindering the widespread adoption of reliable DL architectures in BCIs outside research laboratories. To improve adoption, we need to improve user comfort using, for instance, reliable algorithms that operate with few electrodes. Approach: Our research aims to develop a DL algorithm that delivers effective results with a limited number of electrodes. Taking advantage of the Augmented Covariance Method with SPDNet, we propose the SPDNet$_{\psi}$ architecture and analyze its performance and computational impact, as well as the interpretability of the results. The evaluation is conducted on 5-fold cross-validation, using only three electrodes positioned above the Motor Cortex. The methodology was tested on nearly 100 subjects from several open-source datasets using the Mother Of All BCI Benchmark (MOABB) framework. Main results: The results of our SPDNet$_{\psi}$ demonstrate that the augmented approach combined with the SPDNet significantly outperforms all the current state-of-the-art DL architecture in MI decoding. Significance: This new architecture is explainable, with a low number of trainable parameters and a reduced carbon footprint.
Authors: Yaxi Hu, Amartya Sanyal, Bernhard Sch\"olkopf
Abstract: When analysing Differentially Private (DP) machine learning pipelines, the potential privacy cost of data-dependent pre-processing is frequently overlooked in privacy accounting. In this work, we propose a general framework to evaluate the additional privacy cost incurred by non-private data-dependent pre-processing algorithms. Our framework establishes upper bounds on the overall privacy guarantees by utilising two new technical notions: a variant of DP termed Smooth DP and the bounded sensitivity of the pre-processing algorithms. In addition to the generic framework, we provide explicit overall privacy guarantees for multiple data-dependent pre-processing algorithms, such as data imputation, quantization, deduplication and PCA, when used in combination with several DP algorithms. Notably, this framework is also simple to implement, allowing direct integration into existing DP pipelines.
Authors: Hannah Janmohamed, Marta Wolinska, Shikha Surana, Thomas Pierrot, Aron Walsh, Antoine Cully
Abstract: Crystal structures are indispensable across various domains, from batteries to solar cells, and extensive research has been dedicated to predicting their properties based on their atomic configurations. However, prevailing Crystal Structure Prediction methods focus on identifying the most stable solutions that lie at the global minimum of the energy function. This approach overlooks other potentially interesting materials that lie in neighbouring local minima and have different material properties such as conductivity or resistance to deformation. By contrast, Quality-Diversity algorithms provide a promising avenue for Crystal Structure Prediction as they aim to find a collection of high-performing solutions that have diverse characteristics. However, it may also be valuable to optimise for the stability of crystal structures alongside other objectives such as magnetism or thermoelectric efficiency. Therefore, in this work, we harness the power of Multi-Objective Quality-Diversity algorithms in order to find crystal structures which have diverse features and achieve different trade-offs of objectives. We analyse our approach on 5 crystal systems and demonstrate that it is not only able to re-discover known real-life structures, but also find promising new ones. Moreover, we propose a method for illuminating the objective space to gain an understanding of what trade-offs can be achieved.
Authors: Jincheng Dai, Zhuowei Huang, Haiyun Jiang, Chen Chen, Deng Cai, Wei Bi, Shuming Shi
Abstract: Large Language Models (LLMs), despite their remarkable performance across a wide range of tasks, necessitate substantial GPU memory and consume significant computational resources. Beyond the memory taken up by model weights, the memory used by the KV cache rises linearly with sequence length, becoming a primary bottleneck for inference. In this paper, we introduce an innovative method for optimizing the KV cache, which considerably minimizes its memory footprint. Upon thorough investigation, we discover that in most Transformer models, (i) there is a striking similarity between adjacent tokens' query vectors, and (ii) the attention calculation of the current query can rely exclusively on the attention information of a small fraction of preceding queries. Based on these observations, we present CORM, a KV cache eviction policy that dynamically retains essential key-value pairs for inference without the need for model fine-tuning. Our validation shows that CORM reduces the inference memory usage of KV cache by up to 70\% with negligible performance degradation across six tasks in LongBench. Furthermore, we demonstrate that CORM is compatible with GQA for further compression rate.
Authors: Ben Tu, Nikolas Kantas, Robert M. Lee, Behrang Shafei
Abstract: The goal of multi-objective optimisation is to identify the Pareto front surface which is the set obtained by connecting the best trade-off points. Typically this surface is computed by evaluating the objectives at different points and then interpolating between the subset of the best evaluated trade-off points. In this work, we propose to parameterise the Pareto front surface using polar coordinates. More precisely, we show that any Pareto front surface can be equivalently represented using a scalar-valued length function which returns the projected length along any positive radial direction. We then use this representation in order to rigorously develop the theory and applications of stochastic Pareto front surfaces. In particular, we derive many Pareto front surface statistics of interest such as the expectation, covariance and quantiles. We then discuss how these can be used in practice within a design of experiments setting, where the goal is to both infer and use the Pareto front surface distribution in order to make effective decisions. Our framework allows for clear uncertainty quantification and we also develop advanced visualisation techniques for this purpose. Finally we discuss the applicability of our ideas within multivariate extreme value theory and illustrate our methodology in a variety of numerical examples, including a case study with a real-world air pollution data set.
Authors: Kevin Xin, Lizhi Xin
Abstract: Forecasting, to estimate future events, is crucial for business and decision-making. This paper proposes QxEAI, a methodology that produces a probabilistic forecast that utilizes a quantum-like evolutionary algorithm based on training a quantum-like logic decision tree and a classical value tree on a small number of related time series. We demonstrate how the application of our quantum-like evolutionary algorithm to forecasting can overcome the challenges faced by classical and other machine learning approaches. By using three real-world datasets (Dow Jones Index, retail sales, gas consumption), we show how our methodology produces accurate forecasts while requiring little to none manual work.
Authors: Joongkyu Lee, Min-hwan Oh
Abstract: In this paper, we study the contextual multinomial logit (MNL) bandit problem in which a learning agent sequentially selects an assortment based on contextual information, and user feedback follows an MNL choice model. There has been a significant discrepancy between lower and upper regret bounds, particularly regarding the maximum assortment size $K$. Additionally, the variation in reward structures between these bounds complicates the quest for optimality. Under uniform rewards, where all items have the same expected reward, we establish a regret lower bound of $\Omega(d\sqrt{\smash[b]{T/K}})$ and propose a constant-time algorithm, OFU-MNL+, that achieves a matching upper bound of $\tilde{O}(d\sqrt{\smash[b]{T/K}})$. Under non-uniform rewards, we prove a lower bound of $\Omega(d\sqrt{T})$ and an upper bound of $\tilde{O}(d\sqrt{T})$, also achievable by OFU-MNL+. Our empirical studies support these theoretical findings. To the best of our knowledge, this is the first work in the contextual MNL bandit literature to prove minimax optimality -- for either uniform or non-uniform reward setting -- and to propose a computationally efficient algorithm that achieves this optimality up to logarithmic factors.
Authors: Juyoung Yun, Jungmin Shin
Abstract: Solar flares, especially C, M, and X class, pose significant risks to satellite operations, communication systems, and power grids. We present a novel approach for predicting extreme solar flares using HMI intensitygrams and magnetograms. By detecting sunspots from intensitygrams and extracting magnetic field patches from magnetograms, we train a Residual Network (ResNet) to classify extreme class flares. Our model demonstrates high accuracy, offering a robust tool for predicting extreme solar flares and improving space weather forecasting. Additionally, we show that HMI magnetograms provide more useful data for deep learning compared to other SDO AIA images by better capturing features critical for predicting flare magnitudes. This study underscores the importance of identifying magnetic fields in solar flare prediction, marking a significant advancement in solar activity prediction with practical implications for mitigating space weather impacts.
Authors: Kaixuan Huang, Xudong Guo, Mengdi Wang
Abstract: Speculative decoding reduces the inference latency of a target large language model via utilizing a smaller and faster draft model. Its performance depends on a hyperparameter K -- the candidate length, i.e., the number of candidate tokens for the target model to verify in each round. However, previous methods often use simple heuristics to choose K, which may result in sub-optimal performance. We study the choice of the candidate length K and formulate it as a Markov Decision Process. We theoretically show that the optimal policy of this Markov decision process takes the form of a threshold policy, i.e., the current speculation should stop and be verified when the probability of getting a rejection exceeds a threshold value. Motivated by this theory, we propose SpecDec++, an enhanced version of speculative decoding that adaptively determines the candidate length on the fly. We augment the draft model with a trained acceptance prediction head to predict the conditional acceptance probability of the candidate tokens. SpecDec++ will stop the current speculation when the predicted probability that at least one token gets rejected exceeds a threshold. We implement SpecDec++ and apply it to the llama-2-chat 7B & 70B model pair. Our adaptive method achieves a 2.04x speedup on the Alpaca dataset (an additional 7.2% improvement over the baseline speculative decoding). On the GSM8K and HumanEval datasets, our method achieves a 2.26x speedup (9.4% improvement) and 2.23x speedup (11.1% improvement), respectively.
Authors: Jiesong Lian
Abstract: A popular approach for solving zero-sum games is to maintain populations of policies to approximate the Nash Equilibrium (NE). Previous studies have shown that Policy Space Response Oracle (PSRO) algorithm is an effective multi-agent reinforcement learning framework for solving such games. However, repeatedly training new policies from scratch to approximate Best Response (BR) to opponents' mixed policies at each iteration is both inefficient and costly. While some PSRO variants initialize a new policy by inheriting from past BR policies, this approach limits the exploration of new policies, especially against challenging opponents. To address this issue, we propose Fusion-PSRO, which employs policy fusion to initialize policies for better approximation to BR. By selecting high-quality base policies from meta-NE, policy fusion fuses the base policies into a new policy through model averaging. This approach allows the initialized policies to incorporate multiple expert policies, making it easier to handle difficult opponents compared to inheriting from past BR policies or initializing from scratch. Moreover, our method only modifies the policy initialization phase, allowing its application to nearly all PSRO variants without additional training overhead. Our experiments on non-transitive matrix games, Leduc Poker, and the more complex Liars Dice demonstrate that Fusion-PSRO enhances the performance of nearly all PSRO variants, achieving lower exploitability.
Authors: Sigur de Vries, Sander Keemink, Marcel van Gerven
Abstract: Artificial intelligence (AI) techniques are increasingly being applied to solve control problems. However, control systems developed in AI are often black-box methods, in that it is not clear how and why they generate their outputs. A lack of transparency can be problematic for control tasks in particular, because it complicates the identification of biases or errors, which in turn negatively influences the user's confidence in the system. To improve the interpretability and transparency in control systems, the black-box structure can be replaced with white-box symbolic policies described by mathematical expressions. Genetic programming offers a gradient-free method to optimise the structure of non-differentiable mathematical expressions. In this paper, we show that genetic programming can be used to discover symbolic control systems. This is achieved by learning a symbolic representation of a function that transforms observations into control signals. We consider both systems that implement static control policies without memory and systems that implement dynamic memory-based control policies. In case of the latter, the discovered function becomes the state equation of a differential equation, which allows for evidence integration. Our results show that symbolic policies are discovered that perform comparably with black-box policies on a variety of control tasks. Furthermore, the additional value of the memory capacity in the dynamic policies is demonstrated on experiments where static policies fall short. Overall, we demonstrate that white-box symbolic policies can be optimised with genetic programming, while offering interpretability and transparency that lacks in black-box models.
Authors: Ons Aouedi, Thai-Hoc Vu, Alessio Sacco, Dinh C. Nguyen, Kandaraj Piamrat, Guido Marchetto, Quoc-Viet Pham
Abstract: The rapid advances in the Internet of Things (IoT) have promoted a revolution in communication technology and offered various customer services. Artificial intelligence (AI) techniques have been exploited to facilitate IoT operations and maximize their potential in modern application scenarios. In particular, the convergence of IoT and AI has led to a new networking paradigm called Intelligent IoT (IIoT), which has the potential to significantly transform businesses and industrial domains. This paper presents a comprehensive survey of IIoT by investigating its significant applications in mobile networks, as well as its associated security and privacy issues. Specifically, we explore and discuss the roles of IIoT in a wide range of key application domains, from smart healthcare and smart cities to smart transportation and smart industries. Through such extensive discussions, we investigate important security issues in IIoT networks, where network attacks, confidentiality, integrity, and intrusion are analyzed, along with a discussion of potential countermeasures. Privacy issues in IIoT networks were also surveyed and discussed, including data, location, and model privacy leakage. Finally, we outline several key challenges and highlight potential research directions in this important area.
Authors: Daniel Yun
Abstract: In this paper, we introduce a novel method for merging the weights of multiple pre-trained neural networks using a genetic algorithm called MeGA. Traditional techniques, such as weight averaging and ensemble methods, often fail to fully harness the capabilities of pre-trained networks. Our approach leverages a genetic algorithm with tournament selection, crossover, and mutation to optimize weight combinations, creating a more effective fusion. This technique allows the merged model to inherit advantageous features from both parent models, resulting in enhanced accuracy and robustness. Through experiments on the CIFAR-10 dataset, we demonstrate that our genetic algorithm-based weight merging method improves test accuracy compared to individual models and conventional methods. This approach provides a scalable solution for integrating multiple pre-trained networks across various deep learning applications. Github is available at: https://github.com/YUNBLAK/MeGA-Merging-Multiple-Independently-Trained-Neural-Networks-Based-on-Genetic-Algorithm
Authors: M. J. J. de Grauw, E. Th. Scholten, E. J. Smit, M. J. C. M. Rutten, M. Prokop, B. van Ginneken, A. Hering
Abstract: Size measurements of tumor manifestations on follow-up CT examinations are crucial for evaluating treatment outcomes in cancer patients. Efficient lesion segmentation can speed up these radiological workflows. While numerous benchmarks and challenges address lesion segmentation in specific organs like the liver, kidneys, and lungs, the larger variety of lesion types encountered in clinical practice demands a more universal approach. To address this gap, we introduced the ULS23 benchmark for 3D universal lesion segmentation in chest-abdomen-pelvis CT examinations. The ULS23 training dataset contains 38,693 lesions across this region, including challenging pancreatic, colon and bone lesions. For evaluation purposes, we curated a dataset comprising 775 lesions from 284 patients. Each of these lesions was identified as a target lesion in a clinical context, ensuring diversity and clinical relevance within this dataset. The ULS23 benchmark is publicly accessible via uls23.grand-challenge.org, enabling researchers worldwide to assess the performance of their segmentation methods. Furthermore, we have developed and publicly released our baseline semi-supervised 3D lesion segmentation model. This model achieved an average Dice coefficient of 0.703 $\pm$ 0.240 on the challenge test set. We invite ongoing submissions to advance the development of future ULS models.
Authors: Minghong Fang, Zifan Zhang, Hairi, Prashant Khanduri, Jia Liu, Songtao Lu, Yuchen Liu, Neil Gong
Abstract: Federated learning (FL) enables multiple clients to collaboratively train machine learning models without revealing their private training data. In conventional FL, the system follows the server-assisted architecture (server-assisted FL), where the training process is coordinated by a central server. However, the server-assisted FL framework suffers from poor scalability due to a communication bottleneck at the server, and trust dependency issues. To address challenges, decentralized federated learning (DFL) architecture has been proposed to allow clients to train models collaboratively in a serverless and peer-to-peer manner. However, due to its fully decentralized nature, DFL is highly vulnerable to poisoning attacks, where malicious clients could manipulate the system by sending carefully-crafted local models to their neighboring clients. To date, only a limited number of Byzantine-robust DFL methods have been proposed, most of which are either communication-inefficient or remain vulnerable to advanced poisoning attacks. In this paper, we propose a new algorithm called BALANCE (Byzantine-robust averaging through local similarity in decentralization) to defend against poisoning attacks in DFL. In BALANCE, each client leverages its own local model as a similarity reference to determine if the received model is malicious or benign. We establish the theoretical convergence guarantee for BALANCE under poisoning attacks in both strongly convex and non-convex settings. Furthermore, the convergence rate of BALANCE under poisoning attacks matches those of the state-of-the-art counterparts in Byzantine-free settings. Extensive experiments also demonstrate that BALANCE outperforms existing DFL methods and effectively defends against poisoning attacks.
Authors: A. L. Garc\'ia Navarro, Nataliia Koneva, Alfonso S\'anchez-Maci\'an, Jos\'e Alberto Hern\'andez, \'Oscar Gonz\'alez de Dios, J. M. Rivas-Moscoso
Abstract: This article provides a methodology and open-source implementation of Reinforcement Learning algorithms for finding optimal routes in a packet-optical network scenario. The algorithm uses measurements provided by the physical layer (pre-FEC bit error rate and propagation delay) and the link layer (link load) to configure a set of latency-based rewards and penalties based on such measurements. Then, the algorithm executes Q-learning based on this set of rewards for finding the optimal routing strategies. It is further shown that the algorithm dynamically adapts to changing network conditions by re-calculating optimal policies upon either link load changes or link degradation as measured by pre-FEC BER.
Authors: Siemen Brussee, Giorgio Buzzanca, Anne M. R. Schrader, Jesper Kers
Abstract: Histopathological analysis of Whole Slide Images (WSIs) has seen a surge in the utilization of deep learning methods, particularly Convolutional Neural Networks (CNNs). However, CNNs often fall short in capturing the intricate spatial dependencies inherent in WSIs. Graph Neural Networks (GNNs) present a promising alternative, adept at directly modeling pairwise interactions and effectively discerning the topological tissue and cellular structures within WSIs. Recognizing the pressing need for deep learning techniques that harness the topological structure of WSIs, the application of GNNs in histopathology has experienced rapid growth. In this comprehensive review, we survey GNNs in histopathology, discuss their applications, and explore emerging trends that pave the way for future advancements in the field. We begin by elucidating the fundamentals of GNNs and their potential applications in histopathology. Leveraging quantitative literature analysis, we identify four emerging trends: Hierarchical GNNs, Adaptive Graph Structure Learning, Multimodal GNNs, and Higher-order GNNs. Through an in-depth exploration of these trends, we offer insights into the evolving landscape of GNNs in histopathological analysis. Based on our findings, we propose future directions to propel the field forward. Our analysis serves to guide researchers and practitioners towards innovative approaches and methodologies, fostering advancements in histopathological analysis through the lens of graph neural networks.
Authors: Matthew T. C. Li, Tiangang Cui, Fengyi Li, Youssef Marzouk, Olivier Zahm
Abstract: Identifying low-dimensional structure in high-dimensional probability measures is an essential pre-processing step for efficient sampling. We introduce a method for identifying and approximating a target measure $\pi$ as a perturbation of a given reference measure $\mu$ along a few significant directions of $\mathbb{R}^{d}$. The reference measure can be a Gaussian or a nonlinear transformation of a Gaussian, as commonly arising in generative modeling. Our method extends prior work on minimizing majorizations of the Kullback--Leibler divergence to identify optimal approximations within this class of measures. Our main contribution unveils a connection between the \emph{dimensional} logarithmic Sobolev inequality (LSI) and approximations with this ansatz. Specifically, when the target and reference are both Gaussian, we show that minimizing the dimensional LSI is equivalent to minimizing the KL divergence restricted to this ansatz. For general non-Gaussian measures, the dimensional LSI produces majorants that uniformly improve on previous majorants for gradient-based dimension reduction. We further demonstrate the applicability of this analysis to the squared Hellinger distance, where analogous reasoning shows that the dimensional Poincar\'e inequality offers improved bounds.
Authors: Banibrata Ghosh, Haripriya Harikumar, Khoa D Doan, Svetha Venkatesh, Santu Rana
Abstract: Learning composite concepts, such as \textquotedbl red car\textquotedbl , from individual examples -- like a white car representing the concept of \textquotedbl car\textquotedbl{} and a red strawberry representing the concept of \textquotedbl red\textquotedbl -- is inherently challenging. This paper introduces a novel method called Composite Concept Extractor (CoCE), which leverages techniques from traditional backdoor attacks to learn these composite concepts in a zero-shot setting, requiring only examples of individual concepts. By repurposing the trigger-based model backdooring mechanism, we create a strategic distortion in the manifold of the target object (e.g., \textquotedbl car\textquotedbl ) induced by example objects with the target property (e.g., \textquotedbl red\textquotedbl ) from objects \textquotedbl red strawberry\textquotedbl , ensuring the distortion selectively affects the target objects with the target property. Contrastive learning is then employed to further refine this distortion, and a method is formulated for detecting objects that are influenced by the distortion. Extensive experiments with in-depth analysis across different datasets demonstrate the utility and applicability of our proposed approach.
Authors: Qianhui Wan, Zecheng Zhang, Liheng Jiang, Zhaoqi Wang, Yan Zhou
Abstract: Image anomaly detection is a popular research direction, with many methods emerging in recent years due to rapid advancements in computing. The use of artificial intelligence for image anomaly detection has been widely studied. By analyzing images of athlete posture and movement, it is possible to predict injury status and suggest necessary adjustments. Most existing methods rely on convolutional networks to extract information from irrelevant pixel data, limiting model accuracy. This paper introduces a network combining Residual Network (ResNet) and Bidirectional Gated Recurrent Unit (BiGRU), which can predict potential injury types and provide early warnings by analyzing changes in muscle and bone poses from video images. To address the high complexity of this network, the Sparrow search algorithm was used for optimization. Experiments conducted on four datasets demonstrated that our model has the smallest error in image anomaly detection compared to other models, showing strong adaptability. This provides a new approach for anomaly detection and predictive analysis in images, contributing to the sustainable development of human health and performance.
Authors: Mandar Sharma, Nikhil Muralidhar, Shengzhe Xu, Raquib Bin Yousuf, Naren Ramakrishnan
Abstract: The pretraining-fine-tuning paradigm has been the de facto strategy for transfer learning in modern language modeling. With the understanding that task adaptation in LMs is often a function of parameters shared across tasks, we argue that a more surgical approach to regularization needs to exist for smoother transfer learning. Towards this end, we investigate how the pretraining loss landscape is affected by these task-sensitive parameters through an information-theoretic lens. We then leverage the findings from our investigations to devise a novel approach to dropout for improved model regularization and better downstream generalization. This approach, named guided dropout, is both task & architecture agnostic and adds no computational overhead to the fine-tuning process. Through empirical evaluations, we showcase that our approach to regularization yields consistently better performance, even in scenarios of data paucity, compared to standardized baselines.
Authors: Minsang Kim, Seungjun Baek
Abstract: Compute-efficient training of large language models (LLMs) has become an important research problem. In this work, we consider data pruning as a method of data-efficient training of LLMs, where we take a data compression view on data pruning. We argue that the amount of information of a sample, or the achievable compression on its description length, represents its sample importance. The key idea is that, less informative samples are likely to contain redundant information, and thus should be pruned first. We leverage log-likelihood function of trained models as a surrogate to measure information content of samples. Experiments reveal a surprising insight that information-based pruning can enhance the generalization capability of the model, improves upon language modeling and downstream tasks as compared to the model trained on the entire dataset.
Authors: Xiaohan Lin, Qingxing Cao, Yinya Huang, Haiming Wang, Jianqiao Lu, Zhengying Liu, Linqi Song, Xiaodan Liang
Abstract: Formal verification (FV) has witnessed growing significance with current emerging program synthesis by the evolving large language models (LLMs). However, current formal verification mainly resorts to symbolic verifiers or hand-craft rules, resulting in limitations for extensive and flexible verification. On the other hand, formal languages for automated theorem proving, such as Isabelle, as another line of rigorous verification, are maintained with comprehensive rules and theorems. In this paper, we propose FVEL, an interactive Formal Verification Environment with LLMs. Specifically, FVEL transforms a given code to be verified into Isabelle, and then conducts verification via neural automated theorem proving with an LLM. The joined paradigm leverages the rigorous yet abundant formulated and organized rules in Isabelle and is also convenient for introducing and adjusting cutting-edge LLMs. To achieve this goal, we extract a large-scale FVELER3. The FVELER dataset includes code dependencies and verification processes that are formulated in Isabelle, containing 758 theories, 29,125 lemmas, and 200,646 proof steps in total with in-depth dependencies. We benchmark FVELER in the FVEL environment by first fine-tuning LLMs with FVELER and then evaluating them on Code2Inv and SV-COMP. The results show that FVEL with FVELER fine-tuned Llama3- 8B solves 17.39% (69 -> 81) more problems, and Mistral-7B 12% (75 -> 84) more problems in SV-COMP. And the proportion of proof errors is reduced. Project page: https://fveler.github.io/.