abs
stringlengths 44
64
| Download PDF
stringlengths 75
115
| OpenReview
stringlengths 42
42
| title
stringlengths 15
148
| url
stringlengths 44
64
| authors
stringlengths 6
903
| detail_url
stringlengths 44
64
| tags
stringclasses 1
value | abstract
stringlengths 422
5.84k
|
|---|---|---|---|---|---|---|---|---|
https://proceedings.mlr.press/v235/yan24e.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yan24e/yan24e.pdf
|
https://openreview.net/forum?id=4Zr7T6UrBS
|
Offline Imitation from Observation via Primal Wasserstein State Occupancy Matching
|
https://proceedings.mlr.press/v235/yan24e.html
|
Kai Yan, Alex Schwing, Yu-Xiong Wang
|
https://proceedings.mlr.press/v235/yan24e.html
|
ICML 2024
|
In real-world scenarios, arbitrary interactions with the environment can often be costly, and actions of expert demonstrations are not always available. To reduce the need for both, offline Learning from Observations (LfO) is extensively studied: the agent learns to solve a task given only expert states and task-agnostic non-expert state-action pairs. The state-of-the-art DIstribution Correction Estimation (DICE) methods, as exemplified by SMODICE, minimize the state occupancy divergence between the learner’s and empirical expert policies. However, such methods are limited to either $f$-divergences (KL and $\chi^2$) or Wasserstein distance with Rubinstein duality, the latter of which constrains the underlying distance metric crucial to the performance of Wasserstein-based solutions. To enable more flexible distance metrics, we propose Primal Wasserstein DICE (PW-DICE). It minimizes the primal Wasserstein distance between the learner and expert state occupancies and leverages a contrastively learned distance metric. Theoretically, our framework is a generalization of SMODICE, and is the first work that unifies $f$-divergence and Wasserstein minimization. Empirically, we find that PW-DICE improves upon several state-of-the-art methods. The code is available at https://github.com/KaiYan289/PW-DICE.
|
https://proceedings.mlr.press/v235/yan24f.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yan24f/yan24f.pdf
|
https://openreview.net/forum?id=zWIS8I9G9B
|
Handling Heterogeneous Curvatures in Bandit LQR Control
|
https://proceedings.mlr.press/v235/yan24f.html
|
Yu-Hu Yan, Jing Wang, Peng Zhao
|
https://proceedings.mlr.press/v235/yan24f.html
|
ICML 2024
|
We investigate online Linear Quadratic Regulator (LQR) with bandit feedback and semi-adversarial disturbances. Previous works assume costs with homogeneous curvatures (i.e., with a uniform strong convexity lower bound), which can be hard to satisfy in many real scenarios and prohibits adapting to true curvatures for better performance. In this paper, we initiate the study of bandit LQR control with heterogeneous cost curvatures, aiming to strengthen the algorithm’s adaptivity. To achieve this, we reduce the problem to bandit convex optimization with memory via a “with-history” reduction to avoid hard-to-control truncation errors. Then we provide a novel analysis for an important stability term that appeared in both regret and memory, using Newton decrement developed in interior-point methods. The analysis enables us to guarantee memory-related terms introduced in the reduction and also provide a simplified analysis for handling heterogeneous curvatures in bandit convex optimization. Finally, we achieve interpolated guarantees that can not only recover existing bounds for convex and quadratic costs but also attain new implications for cases of corrupted and decaying quadraticity.
|
https://proceedings.mlr.press/v235/yan24g.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yan24g/yan24g.pdf
|
https://openreview.net/forum?id=oUmXcewb83
|
Foundations of Testing for Finite-Sample Causal Discovery
|
https://proceedings.mlr.press/v235/yan24g.html
|
Tom Yan, Ziyu Xu, Zachary Chase Lipton
|
https://proceedings.mlr.press/v235/yan24g.html
|
ICML 2024
|
Discovery of causal relationships is a fundamental goal of science and vital for sound decision making. As such, there has been considerable interest in causal discovery methods with provable guarantees. Existing works have thus far largely focused on discovery under hard intervention and infinite-samples, in which intervening on a node readily reveals the orientation of every edge incident to the node. This setup however overlooks the stochasticity inherent in real-world, finite-sample settings. Our work takes a step towards studying finite-sample causal discovery, wherein multiple interventions on a node are now needed for edge orientation. In this work, we study the canonical setup in theoretical causal discovery literature, where one assumes causal sufficiency and access to the graph skeleton. Our key observation is that discovery may be viewed as structured, multiple testing, and we develop a novel testing framework to this end. Crucially, our framework allows for anytime valid testing as multiple tests are needed to conclude an edge orientation. It also allows for flexible combination of structured test-statistics (enabling one to use Meek rules to propagate edge orientation) as well as robust testing. Through empirical simulations, we confirm the usefulness of our framework. In closing, using this testing framework, we show how one may efficiently verify graph structure by drawing a connection to multi-constraint bandits and designing a novel algorithm to this end.
|
https://proceedings.mlr.press/v235/yan24h.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yan24h/yan24h.pdf
|
https://openreview.net/forum?id=GVmvBNxB73
|
Retrieval Across Any Domains via Large-scale Pre-trained Model
|
https://proceedings.mlr.press/v235/yan24h.html
|
Jiexi Yan, Zhihui Yin, Chenghao Xu, Cheng Deng, Heng Huang
|
https://proceedings.mlr.press/v235/yan24h.html
|
ICML 2024
|
In order to enhance the generalization ability towards unseen domains, universal cross-domain image retrieval methods require a training dataset encompassing diverse domains, which is costly to assemble. Given this constraint, we introduce a novel problem of data-free adaptive cross-domain retrieval, eliminating the need for real images during training. Towards this goal, we propose a novel Text-driven Knowledge Integration (TKI) method, which exclusively utilizes a pre-trained vision-language model to implement an “aggregation after expansion" training strategy. Specifically, we extract diverse implicit domain-specific information through a set of learnable domain word vectors. Subsequently, a domain-agnostic universal projection, equipped with a non-Euclidean multi-layer perceptron, can be optimized using these assorted text descriptions through the text-proxied domain aggregation. Leveraging the cross-modal transferability phenomenon of the shared latent space, we can integrate the trained domain-agnostic universal projection with the pre-trained visual encoder to extract the features of the input image for the following retrieval during testing. Extensive experimental results on several benchmark datasets demonstrate the superiority of our method.
|
https://proceedings.mlr.press/v235/yan24i.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yan24i/yan24i.pdf
|
https://openreview.net/forum?id=GktjBAGgo4
|
Reducing Balancing Error for Causal Inference via Optimal Transport
|
https://proceedings.mlr.press/v235/yan24i.html
|
Yuguang Yan, Hao Zhou, Zeqin Yang, Weilin Chen, Ruichu Cai, Zhifeng Hao
|
https://proceedings.mlr.press/v235/yan24i.html
|
ICML 2024
|
Most studies on causal inference tackle the issue of confounding bias by reducing the distribution shift between the control and treated groups. However, it remains an open question to adopt an appropriate metric for distribution shift in practice. In this paper, we define a generic balancing error on reweighted samples to characterize the confounding bias, and study the connection between the balancing error and the Wasserstein discrepancy derived from the theory of optimal transport. We not only regard the Wasserstein discrepancy as the metric of distribution shift, but also explore the association between the balancing error and the underlying cost function involved in the Wasserstein discrepancy. Motivated by this, we propose to reduce the balancing error under the framework of optimal transport with learnable marginal distributions and the cost function, which is implemented by jointly learning weights and representations associated with factual outcomes. The experiments on both synthetic and real-world datasets demonstrate the effectiveness of our proposed method.
|
https://proceedings.mlr.press/v235/yang24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24a/yang24a.pdf
|
https://openreview.net/forum?id=xLikRS9OhW
|
Do Efficient Transformers Really Save Computation?
|
https://proceedings.mlr.press/v235/yang24a.html
|
Kai Yang, Jan Ackermann, Zhenyu He, Guhao Feng, Bohang Zhang, Yunzhen Feng, Qiwei Ye, Di He, Liwei Wang
|
https://proceedings.mlr.press/v235/yang24a.html
|
ICML 2024
|
As transformer-based language models are trained on increasingly large datasets and with vast numbers of parameters, finding more efficient alternatives to the standard Transformer has become very valuable. While many efficient Transformers and Transformer alternatives have been proposed, none provide theoretical guarantees that they are a suitable replacement for the standard Transformer. This makes it challenging to identify when to use a specific model and what directions to prioritize for further investigation. In this paper, we aim to understand the capabilities and limitations of efficient Transformers, specifically the Sparse Transformer and the Linear Transformer. We focus on their reasoning capability as exhibited by Chain-of-Thought (CoT) prompts and follow previous works to model them as Dynamic Programming (DP) problems. Our results show that while these models are expressive enough to solve general DP tasks, contrary to expectations, they require a model size that scales with the problem size. Nonetheless, we identify a class of DP problems for which these models can be more efficient than the standard Transformer. We confirm our theoretical results through experiments on representative DP tasks, adding to the understanding of efficient Transformers’ practical strengths and weaknesses.
|
https://proceedings.mlr.press/v235/yang24b.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24b/yang24b.pdf
|
https://openreview.net/forum?id=hWng0GXeE4
|
Mind the Boundary: Coreset Selection via Reconstructing the Decision Boundary
|
https://proceedings.mlr.press/v235/yang24b.html
|
Shuo Yang, Zhe Cao, Sheng Guo, Ruiheng Zhang, Ping Luo, Shengping Zhang, Liqiang Nie
|
https://proceedings.mlr.press/v235/yang24b.html
|
ICML 2024
|
Existing paradigms of pushing the state of the art require exponentially more training data in many fields. Coreset selection seeks to mitigate this growing demand by identifying the most efficient subset of training data. In this paper, we delve into geometry-based coreset methods and preliminarily link the geometry of data distribution with models’ generalization capability in theoretics. Leveraging these theoretical insights, we propose a novel coreset construction method by selecting training samples to reconstruct the decision boundary of a deep neural network learned on the full dataset. Extensive experiments across various popular benchmarks demonstrate the superiority of our method over multiple competitors. For the first time, our method achieves a 50% data pruning rate on the ImageNet-1K dataset while sacrificing less than 1% in accuracy. Additionally, we showcase and analyze the remarkable cross-architecture transferability of the coresets derived from our approach.
|
https://proceedings.mlr.press/v235/yang24c.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24c/yang24c.pdf
|
https://openreview.net/forum?id=w8ei1o9U5y
|
Sample-Efficient Multiagent Reinforcement Learning with Reset Replay
|
https://proceedings.mlr.press/v235/yang24c.html
|
Yaodong Yang, Guangyong Chen, Jianye Hao, Pheng-Ann Heng
|
https://proceedings.mlr.press/v235/yang24c.html
|
ICML 2024
|
The popularity of multiagent reinforcement learning (MARL) is growing rapidly with the demand for real-world tasks that require swarm intelligence. However, a noticeable drawback of MARL is its low sample efficiency, which leads to a huge amount of interactions with the environment. Surprisingly, few MARL works focus on this practical problem especially in the parallel environment setting, which greatly hampers the application of MARL into the real world. In response to this gap, in this paper, we propose Multiagent Reinforcement Learning with Reset Replay (MARR) to greatly improve the sample efficiency of MARL by enabling MARL training at a high replay ratio in the parallel environment setting for the first time. To achieve this, first, a reset strategy is introduced for maintaining the network plasticity to ensure that MARL continually learns with a high replay ratio. Second, MARR incorporates a data augmentation technique to boost the sample efficiency further. Extensive experiments in SMAC and MPE show that MARR significantly improves the performance of various MARL approaches with much fewer environment interactions.
|
https://proceedings.mlr.press/v235/yang24d.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24d/yang24d.pdf
|
https://openreview.net/forum?id=QMy2RLnxGN
|
DoraemonGPT: Toward Understanding Dynamic Scenes with Large Language Models (Exemplified as A Video Agent)
|
https://proceedings.mlr.press/v235/yang24d.html
|
Zongxin Yang, Guikun Chen, Xiaodi Li, Wenguan Wang, Yi Yang
|
https://proceedings.mlr.press/v235/yang24d.html
|
ICML 2024
|
Recent LLM-driven visual agents mainly focus on solving image-based tasks, which limits their ability to understand dynamic scenes, making it far from real-life applications like guiding students in laboratory experiments and identifying their mistakes. Hence, this paper explores DoraemonGPT, a comprehensive and conceptually elegant system driven by LLMs to understand dynamic scenes. Considering the video modality better reflects the ever-changing nature of real-world scenarios, we exemplify DoraemonGPT as a video agent. Given a video with a question/task, DoraemonGPT begins by converting the input video into a symbolic memory that stores task-related attributes. This structured representation allows for spatial-temporal querying and reasoning by well-designed sub-task tools, resulting in concise intermediate results. Recognizing that LLMs have limited internal knowledge when it comes to specialized domains (e.g., analyzing the scientific principles underlying experiments), we incorporate plug-and-play tools to assess external knowledge and address tasks across different domains. Moreover, a novel LLM-driven planner based on Monte Carlo Tree Search is introduced to explore the large planning space for scheduling various tools. The planner iteratively finds feasible solutions by backpropagating the result’s reward, and multiple solutions can be summarized into an improved final answer. We extensively evaluate DoraemonGPT’s effectiveness on three benchmarks and several in-the-wild scenarios. Project page: https://z-x-yang.github.io/doraemon-gpt.
|
https://proceedings.mlr.press/v235/yang24e.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24e/yang24e.pdf
|
https://openreview.net/forum?id=xVXnXk9I3I
|
A Dense Reward View on Aligning Text-to-Image Diffusion with Preference
|
https://proceedings.mlr.press/v235/yang24e.html
|
Shentao Yang, Tianqi Chen, Mingyuan Zhou
|
https://proceedings.mlr.press/v235/yang24e.html
|
ICML 2024
|
Aligning text-to-image diffusion model (T2I) with preference has been gaining increasing research attention. While prior works exist on directly optimizing T2I by preference data, these methods are developed under the bandit assumption of a latent reward on the entire diffusion reverse chain, while ignoring the sequential nature of the generation process. This may harm the efficacy and efficiency of preference alignment. In this paper, we take on a finer dense reward perspective and derive a tractable alignment objective that emphasizes the initial steps of the T2I reverse chain. In particular, we introduce temporal discounting into DPO-style explicit-reward-free objectives, to break the temporal symmetry therein and suit the T2I generation hierarchy. In experiments on single and multiple prompt generation, our method is competitive with strong relevant baselines, both quantitatively and qualitatively. Further investigations are conducted to illustrate the insight of our approach. Source code is available at https://github.com/Shentao-YANG/Dense_Reward_T2I .
|
https://proceedings.mlr.press/v235/yang24f.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24f/yang24f.pdf
|
https://openreview.net/forum?id=3xPMW9JURD
|
Lyapunov-stable Neural Control for State and Output Feedback: A Novel Formulation
|
https://proceedings.mlr.press/v235/yang24f.html
|
Lujie Yang, Hongkai Dai, Zhouxing Shi, Cho-Jui Hsieh, Russ Tedrake, Huan Zhang
|
https://proceedings.mlr.press/v235/yang24f.html
|
ICML 2024
|
Learning-based neural-network (NN) control policies have shown impressive empirical performance in a wide range of tasks in robotics and control. However, formal (Lyapunov) stability guarantees over the region-of-attraction (ROA) for NN controllers with nonlinear dynamical systems are challenging to obtain, and most existing approaches rely on expensive solvers for sums-of-squares (SOS), mixed-integer programming (MIP), or satisfiability modulo theories (SMT). In this paper, we demonstrate a new framework for learning NN controllers together with Lyapunov certificates using fast empirical falsification and strategic regularizations. We propose a novel formulation that defines a larger verifiable region-of-attraction (ROA) than shown in the literature, and refines the conventional restrictive constraints on Lyapunov derivatives to focus only on certifiable ROAs. The Lyapunov condition is rigorously verified post-hoc using branch-and-bound with scalable linear bound propagation-based NN verification techniques. The approach is efficient and flexible, and the full training and verification procedure is accelerated on GPUs without relying on expensive solvers for SOS, MIP, nor SMT. The flexibility and efficiency of our framework allow us to demonstrate Lyapunov-stable output feedback control with synthesized NN-based controllers and NN-based observers with formal stability guarantees, for the first time in literature.
|
https://proceedings.mlr.press/v235/yang24g.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24g/yang24g.pdf
|
https://openreview.net/forum?id=qstt2OguvM
|
Latent Space Symmetry Discovery
|
https://proceedings.mlr.press/v235/yang24g.html
|
Jianke Yang, Nima Dehmamy, Robin Walters, Rose Yu
|
https://proceedings.mlr.press/v235/yang24g.html
|
ICML 2024
|
Equivariant neural networks require explicit knowledge of the symmetry group. Automatic symmetry discovery methods aim to relax this constraint and learn invariance and equivariance from data. However, existing symmetry discovery methods are limited to simple linear symmetries and cannot handle the complexity of real-world data. We propose a novel generative model, Latent LieGAN (LaLiGAN), which can discover symmetries of nonlinear group actions. It learns a mapping from the data space to a latent space where the symmetries become linear and simultaneously discovers symmetries in the latent space. Theoretically, we show that our model can express nonlinear symmetries under some conditions about the group action. Experimentally, we demonstrate that our method can accurately discover the intrinsic symmetry in high-dimensional dynamical systems. LaLiGAN also results in a well-structured latent space that is useful for downstream tasks including equation discovery and long-term forecasting.
|
https://proceedings.mlr.press/v235/yang24h.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24h/yang24h.pdf
|
https://openreview.net/forum?id=VtqyurB4Af
|
Guidance with Spherical Gaussian Constraint for Conditional Diffusion
|
https://proceedings.mlr.press/v235/yang24h.html
|
Lingxiao Yang, Shutong Ding, Yifan Cai, Jingyi Yu, Jingya Wang, Ye Shi
|
https://proceedings.mlr.press/v235/yang24h.html
|
ICML 2024
|
Recent advances in diffusion models attempt to handle conditional generative tasks by utilizing a differentiable loss function for guidance without the need for additional training. While these methods achieved certain success, they often compromise on sample quality and require small guidance step sizes, leading to longer sampling processes. This paper reveals that the fundamental issue lies in the manifold deviation during the sampling process when loss guidance is employed. We theoretically show the existence of manifold deviation by establishing a certain lower bound for the estimation error of the loss guidance. To mitigate this problem, we propose Diffusion with Spherical Gaussian constraint (DSG), drawing inspiration from the concentration phenomenon in high-dimensional Gaussian distributions. DSG effectively constrains the guidance step within the intermediate data manifold through optimization and enables the use of larger guidance steps. Furthermore, we present a closed-form solution for DSG denoising with the Spherical Gaussian constraint. Notably, DSG can seamlessly integrate as a plugin module within existing training-free conditional diffusion methods. Implementing DSG merely involves a few lines of additional code with almost no extra computational overhead, yet it leads to significant performance improvements. Comprehensive experimental results in various conditional generation tasks validate the superiority and adaptability of DSG in terms of both sample quality and time efficiency.
|
https://proceedings.mlr.press/v235/yang24i.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24i/yang24i.pdf
|
https://openreview.net/forum?id=ycLHJuLYuD
|
Better Safe than Sorry: Pre-training CLIP against Targeted Data Poisoning and Backdoor Attacks
|
https://proceedings.mlr.press/v235/yang24i.html
|
Wenhan Yang, Jingdong Gao, Baharan Mirzasoleiman
|
https://proceedings.mlr.press/v235/yang24i.html
|
ICML 2024
|
Contrastive Language-Image Pre-training (CLIP) on large image-caption datasets has achieved remarkable success in zero-shot classification and enabled transferability to new domains. However, CLIP is extremely more vulnerable to targeted data poisoning and backdoor attacks compared to supervised learning. Perhaps surprisingly, poisoning 0.0001% of CLIP pre-training data is enough to make targeted data poisoning attacks successful. This is four orders of magnitude smaller than what is required to poison supervised models. Despite this vulnerability, existing methods are very limited in defending CLIP models during pre-training. In this work, we propose a strong defense, SAFECLIP, to safely pre-train CLIP against targeted data poisoning and backdoor attacks. SAFECLIP warms up the model by applying unimodal contrastive learning (CL) on image and text modalities separately. Then, it divides the data into safe and risky sets by applying a Gaussian Mixture Model to the cosine similarity of image-caption pair representations. SAFECLIP pre-trains the model by applying the CLIP loss to the safe set and applying unimodal CL to image and text modalities of the risky set separately. By gradually increasing the size of the safe set during pre-training, SAFECLIP effectively breaks targeted data poisoning and backdoor attacks without harming the CLIP performance. Our extensive experiments on CC3M, Visual Genome, and MSCOCO demonstrate that SAFECLIP significantly reduces the success rate of targeted data poisoning attacks from 93.75% to 0% and that of various backdoor attacks from up to 100% to 0%, without harming CLIP’s performance.
|
https://proceedings.mlr.press/v235/yang24j.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24j/yang24j.pdf
|
https://openreview.net/forum?id=IWi6iLZeRG
|
Deeper or Wider: A Perspective from Optimal Generalization Error with Sobolev Loss
|
https://proceedings.mlr.press/v235/yang24j.html
|
Yahong Yang, Juncai He
|
https://proceedings.mlr.press/v235/yang24j.html
|
ICML 2024
|
Constructing the architecture of a neural network is a challenging pursuit for the machine learning community, and the dilemma of whether to go deeper or wider remains a persistent question. This paper explores a comparison between deeper neural networks (DeNNs) with a flexible number of layers and wider neural networks (WeNNs) with limited hidden layers, focusing on their optimal generalization error in Sobolev losses. Analytical investigations reveal that the architecture of a neural network can be significantly influenced by various factors, including the number of sample points, parameters within the neural networks, and the regularity of the loss function. Specifically, a higher number of parameters tends to favor WeNNs, while an increased number of sample points and greater regularity in the loss function lean towards the adoption of DeNNs. We ultimately apply this theory to address partial differential equations using deep Ritz and physics-informed neural network (PINN) methods, guiding the design of neural networks.
|
https://proceedings.mlr.press/v235/yang24k.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24k/yang24k.pdf
|
https://openreview.net/forum?id=M5kn9NKIs4
|
SAM as the Guide: Mastering Pseudo-Label Refinement in Semi-Supervised Referring Expression Segmentation
|
https://proceedings.mlr.press/v235/yang24k.html
|
Danni Yang, Jiayi Ji, Yiwei Ma, Tianyu Guo, Haowei Wang, Xiaoshuai Sun, Rongrong Ji
|
https://proceedings.mlr.press/v235/yang24k.html
|
ICML 2024
|
In this paper, we introduce SemiRES, a semi-supervised framework that effectively leverages a combination of labeled and unlabeled data to perform RES. A significant hurdle in applying semi-supervised techniques to RES is the prevalence of noisy pseudo-labels, particularly at the boundaries of objects. SemiRES incorporates the Segment Anything Model (SAM), renowned for its precise boundary demarcation, to improve the accuracy of these pseudo-labels. Within SemiRES, we offer two alternative matching strategies: IoU-based Optimal Matching (IOM) and Composite Parts Integration (CPI). These strategies are designed to extract the most accurate masks from SAM’s output, thus guiding the training of the student model with enhanced precision. In instances where a precise mask cannot be matched from the available candidates, we develop the Pixel-Wise Adjustment (PWA) strategy, guiding the student model’s training directly by the pseudo-labels. Extensive experiments on three RES benchmarks—RefCOCO, RefCOCO+, and G-Ref reveal its superior performance compared to fully supervised methods, especially in low-data scenarios. Remarkably, with only 1% labeled data, our SemiRES outperforms the supervised baseline by a large margin, e.g. +18.64% gains on RefCOCO val set.
|
https://proceedings.mlr.press/v235/yang24l.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24l/yang24l.pdf
|
https://openreview.net/forum?id=gDQuupz8mm
|
Small-loss Adaptive Regret for Online Convex Optimization
|
https://proceedings.mlr.press/v235/yang24l.html
|
Wenhao Yang, Wei Jiang, Yibo Wang, Ping Yang, Yao Hu, Lijun Zhang
|
https://proceedings.mlr.press/v235/yang24l.html
|
ICML 2024
|
To deal with changing environments, adaptive regret has been proposed to minimize the regret over every interval. Previous studies have established a small-loss adaptive regret bound for general convex functions under the smoothness condition, offering the advantage of being much tighter than minimax rates for benign problems. However, it remains unclear whether similar bounds are attainable for other types of convex functions, such as exp-concave and strongly convex functions. In this paper, we first propose a novel algorithm that achieves a small-loss adaptive regret bound for exp-concave and smooth function. Subsequently, to address the limitation that existing algorithms can only handle one type of convex functions, we further design a universal algorithm capable of delivering small-loss adaptive regret bounds for general convex, exp-concave, and strongly convex functions simultaneously. That is challenging because the universal algorithm follows the meta-expert framework, and we need to ensure that upper bounds for both meta-regret and expert-regret are of small-loss types. Moreover, we provide a novel analysis demonstrating that our algorithms are also equipped with minimax adaptive regret bounds when functions are non-smooth.
|
https://proceedings.mlr.press/v235/yang24m.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24m/yang24m.pdf
|
https://openreview.net/forum?id=1ZJLNLZIpk
|
Towards Interpretable Deep Local Learning with Successive Gradient Reconciliation
|
https://proceedings.mlr.press/v235/yang24m.html
|
Yibo Yang, Xiaojie Li, Motasem Alfarra, Hasan Abed Al Kader Hammoud, Adel Bibi, Philip Torr, Bernard Ghanem
|
https://proceedings.mlr.press/v235/yang24m.html
|
ICML 2024
|
Relieving the reliance of neural network training on a global back-propagation (BP) has emerged as a notable research topic due to the biological implausibility and huge memory consumption caused by BP. Among the existing solutions, local learning optimizes gradient-isolated modules of a neural network with local errors and has been proved to be effective even on large-scale datasets. However, the reconciliation among local errors has never been investigated. In this paper, we first theoretically study non-greedy layer-wise training and show that the convergence cannot be assured when the local gradient in a module w.r.t. its input is not reconciled with the local gradient in the previous module w.r.t. its output. Inspired by the theoretical result, we further propose a local training strategy that successively regularizes the gradient reconciliation between neighboring modules without breaking gradient isolation or introducing any learnable parameters. Our method can be integrated into both local-BP and BP-free settings. In experiments, we achieve significant performance improvements compared to previous methods. Particularly, our method for CNN and Transformer architectures on ImageNet is able to attain a competitive performance with global BP, saving more than 40% memory consumption.
|
https://proceedings.mlr.press/v235/yang24n.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24n/yang24n.pdf
|
https://openreview.net/forum?id=mjh7AOWozN
|
Bounded and Uniform Energy-based Out-of-distribution Detection for Graphs
|
https://proceedings.mlr.press/v235/yang24n.html
|
Shenzhi Yang, Bin Liang, An Liu, Lin Gui, Xingkai Yao, Xiaofang Zhang
|
https://proceedings.mlr.press/v235/yang24n.html
|
ICML 2024
|
Given the critical role of graphs in real-world applications and their high-security requirements, improving the ability of graph neural networks (GNNs) to detect out-of-distribution (OOD) data is an urgent research problem. The recent work GNNSAFE proposes a framework based on the aggregation of negative energy scores that significantly improves the performance of GNNs to detect node-level OOD data. However, our study finds that score aggregation among nodes is susceptible to extreme values due to the unboundedness of the negative energy scores and logit shifts, which severely limits the accuracy of GNNs in detecting node-level OOD data. In this paper, we propose NODESAFE: reducing the generation of extreme scores of nodes by adding two optimization terms that make the negative energy scores bounded and mitigate the logit shift. Experimental results show that our approach dramatically improves the ability of GNNs to detect OOD data at the node level, e.g., in detecting OOD data induced by Structure Manipulation, the metric of FPR95 (lower is better) in scenarios without (with) OOD data exposure are reduced from the current SOTA by 28.4% ( 22.7% ). The code is available via https://github.com/ShenzhiYang2000/NODESAFE-Bounded-and-Uniform-Energy-based-Out-of-distribution-Detection-for-Graphs.
|
https://proceedings.mlr.press/v235/yang24o.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24o/yang24o.pdf
|
https://openreview.net/forum?id=H9fNj8ivTy
|
Position: Towards Implicit Prompt For Text-To-Image Models
|
https://proceedings.mlr.press/v235/yang24o.html
|
Yue Yang, Yuqi Lin, Hong Liu, Wenqi Shao, Runjian Chen, Hailong Shang, Yu Wang, Yu Qiao, Kaipeng Zhang, Ping Luo
|
https://proceedings.mlr.press/v235/yang24o.html
|
ICML 2024
|
Recent text-to-image (T2I) models have had great success, and many benchmarks have been proposed to evaluate their performance and safety. However, they only consider explicit prompts while neglecting implicit prompts (hint at a target without explicitly mentioning it). These prompts may get rid of safety constraints and pose potential threats to the applications of these models. This position paper highlights the current state of T2I models toward implicit prompts. We present a benchmark named ImplicitBench and conduct an investigation on the performance and impacts of implicit prompts with popular T2I models. Specifically, we design and collect more than 2,000 implicit prompts of three aspects: General Symbols, Celebrity Privacy, and Not-Safe-For-Work (NSFW) Issues, and evaluate six well-known T2I models’ capabilities under these implicit prompts. Experiment results show that (1) T2I models are able to accurately create various target symbols indicated by implicit prompts; (2) Implicit prompts bring potential risks of privacy leakage for T2I models. (3) Constraints of NSFW in most of the evaluated T2I models can be bypassed with implicit prompts. We call for increased attention to the potential and risks of implicit prompts in the T2I community and further investigation into the capabilities and impacts of implicit prompts, advocating for a balanced approach that harnesses their benefits while mitigating their risks.
|
https://proceedings.mlr.press/v235/yang24p.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24p/yang24p.pdf
|
https://openreview.net/forum?id=DzLna0cFL1
|
Position: Towards Unified Alignment Between Agents, Humans, and Environment
|
https://proceedings.mlr.press/v235/yang24p.html
|
Zonghan Yang, An Liu, Zijun Liu, Kaiming Liu, Fangzhou Xiong, Yile Wang, Zeyuan Yang, Qingyuan Hu, Xinrui Chen, Zhenhe Zhang, Fuwen Luo, Zhicheng Guo, Peng Li, Yang Liu
|
https://proceedings.mlr.press/v235/yang24p.html
|
ICML 2024
|
The rapid progress of foundation models has led to the prosperity of autonomous agents, which leverage the universal capabilities of foundation models to conduct reasoning, decision-making, and environmental interaction. However, the efficacy of agents remains limited when operating in intricate, realistic environments. In this work, we introduce the principles of Unified Alignment for Agents (UA$^2$), which advocate for the simultaneous alignment of agents with human intentions, environmental dynamics, and self-constraints such as the limitation of monetary budgets. From the perspective of UA$^2$, we review the current agent research and highlight the neglected factors in existing agent benchmarks and method candidates. We also conduct proof-of-concept studies by introducing realistic features to WebShop, including user profiles demonstrating intentions, personalized reranking reflecting complex environmental dynamics, and runtime cost statistics as self-constraints. We then follow the principles of UA$^2$ to propose an initial design of our agent and benchmark its performance with several candidate baselines in the retrofitted WebShop. The extensive experimental results further prove the importance of the principles of UA$^2$. Our research sheds light on the next steps of autonomous agent research with improved general problem-solving abilities.
|
https://proceedings.mlr.press/v235/yang24q.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24q/yang24q.pdf
|
https://openreview.net/forum?id=QLcBzRI3V3
|
Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment
|
https://proceedings.mlr.press/v235/yang24q.html
|
Rui Yang, Xiaoman Pan, Feng Luo, Shuang Qiu, Han Zhong, Dong Yu, Jianshu Chen
|
https://proceedings.mlr.press/v235/yang24q.html
|
ICML 2024
|
We consider the problem of multi-objective alignment of foundation models with human preferences, which is a critical step towards helpful and harmless AI systems. However, it is generally costly and unstable to fine-tune large foundation models using reinforcement learning (RL), and the multi-dimensionality, heterogeneity, and conflicting nature of human preferences further complicate the alignment process. In this paper, we introduce Rewards-in-Context (RiC), which conditions the response of a foundation model on multiple rewards in its prompt context and applies supervised fine-tuning for alignment. The salient features of RiC are simplicity and adaptivity, as it only requires supervised fine-tuning of a single foundation model and supports dynamic adjustment for user preferences during inference time. Inspired by the analytical solution of an abstracted convex optimization problem, our dynamic inference-time adjustment method approaches the Pareto-optimal solution for multiple objectives. Empirical evidence demonstrates the efficacy of our method in aligning both Large Language Models (LLMs) and diffusion models to accommodate diverse rewards with only around 10% GPU hours compared with multi-objective RL baseline.
|
https://proceedings.mlr.press/v235/yang24r.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24r/yang24r.pdf
|
https://openreview.net/forum?id=zwUEk9WpsR
|
Understanding Server-Assisted Federated Learning in the Presence of Incomplete Client Participation
|
https://proceedings.mlr.press/v235/yang24r.html
|
Haibo Yang, Peiwen Qiu, Prashant Khanduri, Minghong Fang, Jia Liu
|
https://proceedings.mlr.press/v235/yang24r.html
|
ICML 2024
|
Existing works in federated learning (FL) often assume either full client or uniformly distributed client participation. However, in reality, some clients may never participate in FL training (aka incomplete client participation) due to various system heterogeneity factors. A popular solution is the server-assisted federated learning (SA-FL) framework, where the server uses an auxiliary dataset. Despite empirical evidence of SA-FL’s effectiveness in addressing incomplete client participation, theoretical understanding of SA-FL is lacking. Furthermore, the effects of incomplete client participation in conventional FL are poorly understood. This motivates us to rigorously investigate SA-FL. Toward this end, we first show that conventional FL is not PAC-learnable under incomplete client participation in the worst case. Then, we show that the PAC-learnability of FL with incomplete client participation can indeed be revived by SA-FL, which theoretically justifies the use of SA-FL for the first time. Lastly, to provide practical guidance for SA-FL training under incomplete client participation, we propose the SAFARI (server-assisted federated averaging) algorithm that enjoys the same linear convergence speedup guarantees as classic FL with ideal client participation assumptions, offering the first SA-FL algorithm with convergence guarantee. Extensive experiments on different datasets show SAFARI significantly improves the performance under incomplete client participation.
|
https://proceedings.mlr.press/v235/yang24s.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24s/yang24s.pdf
|
https://openreview.net/forum?id=b0lxGL2n3d
|
ILILT: Implicit Learning of Inverse Lithography Technologies
|
https://proceedings.mlr.press/v235/yang24s.html
|
Haoyu Yang, Haoxing Ren
|
https://proceedings.mlr.press/v235/yang24s.html
|
ICML 2024
|
Lithography, transferring chip design masks to the silicon wafer, is the most important phase in modern semiconductor manufacturing flow. Due to the limitations of lithography systems, Extensive design optimizations are required to tackle the design and silicon mismatch. Inverse lithography technology (ILT) is one of the promising solutions to perform pre-fabrication optimization, termed mask optimization. Because of mask optimization problems’ constrained non-convexity, numerical ILT solvers rely heavily on good initialization to avoid getting stuck on sub-optimal solutions. Machine learning (ML) techniques are hence proposed to generate mask initialization for ILT solvers with one-shot inference, targeting faster and better convergence during ILT. This paper addresses the question of whether ML models can directly generate high-quality optimized masks without engaging ILT solvers in the loop. We propose an implicit learning ILT framework: ILILT, which leverages the implicit layer learning method and lithography-conditioned inputs to ground the model. Trained to understand the ILT optimization procedure, ILILT can outperform the state-of-the-art machine learning solutions, significantly improving efficiency and quality.
|
https://proceedings.mlr.press/v235/yang24t.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24t/yang24t.pdf
|
https://openreview.net/forum?id=Sbl2keQEML
|
Representation Surgery for Multi-Task Model Merging
|
https://proceedings.mlr.press/v235/yang24t.html
|
Enneng Yang, Li Shen, Zhenyi Wang, Guibing Guo, Xiaojun Chen, Xingwei Wang, Dacheng Tao
|
https://proceedings.mlr.press/v235/yang24t.html
|
ICML 2024
|
Multi-task learning (MTL) compresses the information from multiple tasks into a unified backbone to improve computational efficiency and generalization. Recent work directly merges multiple independently trained models to perform MTL instead of collecting their raw data for joint training, greatly expanding the application scenarios of MTL. However, by visualizing the representation distribution of existing model merging schemes, we find that the merged model often suffers from the dilemma of representation bias. That is, there is a significant discrepancy in the representation distribution between the merged and individual models, resulting in poor performance of merged MTL. In this paper, we propose a representation surgery solution called “Surgery" to reduce representation bias in the merged model. Specifically, Surgery is a lightweight task-specific plugin that takes the representation of the merged model as input and attempts to output the biases contained in the representation from the merged model. We then designed an unsupervised optimization objective that updates the Surgery plugin by minimizing the distance between the merged model’s representation and the individual model’s representation. Extensive experiments demonstrate significant MTL performance improvements when our Surgery plugin is applied to state-of-the-art (SOTA) model merging schemes.
|
https://proceedings.mlr.press/v235/yang24u.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24u/yang24u.pdf
|
https://openreview.net/forum?id=IpSKpOY2EH
|
Reducing Fine-Tuning Memory Overhead by Approximate and Memory-Sharing Backpropagation
|
https://proceedings.mlr.press/v235/yang24u.html
|
Yuchen Yang, Yingdong Shi, Cheems Wang, Xiantong Zhen, Yuxuan Shi, Jun Xu
|
https://proceedings.mlr.press/v235/yang24u.html
|
ICML 2024
|
Fine-tuning pretrained large models to downstream tasks is an important problem, which however suffers from huge memory overhead due to large-scale parameters. This work strives to reduce memory overhead in fine-tuning from perspectives of activation function and layer normalization. To this end, we propose the Approximate Backpropagation (Approx-BP) theory, which provides the theoretical feasibility of decoupling the forward and backward passes. We apply our Approx-BP theory to backpropagation training and derive memory-efficient alternatives of GELU and SiLU activation functions, which use derivative functions of ReLUs in the backward pass while keeping their forward pass unchanged. In addition, we introduce a Memory-Sharing Backpropagation strategy, which enables the activation memory to be shared by two adjacent layers, thereby removing activation memory usage redundancy. Our method neither induces extra computation nor reduces training efficiency. We conduct extensive experiments with pretrained vision and language models, and the results demonstrate that our proposal can reduce up to $\sim$$30%$ of the peak memory usage. Our code is released at github.
|
https://proceedings.mlr.press/v235/yang24v.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24v/yang24v.pdf
|
https://openreview.net/forum?id=87ZrVHDqmR
|
Unlocking the Power of Spatial and Temporal Information in Medical Multimodal Pre-training
|
https://proceedings.mlr.press/v235/yang24v.html
|
Jinxia Yang, Bing Su, Xin Zhao, Ji-Rong Wen
|
https://proceedings.mlr.press/v235/yang24v.html
|
ICML 2024
|
Medical vision-language pre-training methods mainly leverage the correspondence between paired medical images and radiological reports. Although multi-view spatial images and temporal sequences of image-report pairs are available in off-the-shelf multi-modal medical datasets, most existing methods have not thoroughly tapped into such extensive supervision signals. In this paper, we introduce the Med-ST framework for fine-grained spatial and temporal modeling to exploit information from multiple spatial views of chest radiographs and temporal historical records. For spatial modeling, Med-ST employs the Mixture of View Expert (MoVE) architecture to integrate different visual features from both frontal and lateral views. To achieve a more comprehensive alignment, Med-ST not only establishes the global alignment between whole images and texts but also introduces modality-weighted local alignment between text tokens and spatial regions of images. For temporal modeling, we propose a novel cross-modal bidirectional cycle consistency objective by forward mapping classification (FMC) and reverse mapping regression (RMR). By perceiving temporal information from simple to complex, Med-ST can learn temporal semantics. Experimental results across four distinct tasks demonstrate the effectiveness of Med-ST, especially in temporal classification tasks. Our code and model are available at https://github.com/SVT-Yang/MedST.
|
https://proceedings.mlr.press/v235/yang24w.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24w/yang24w.pdf
|
https://openreview.net/forum?id=rIrpzmqRBk
|
Exploration and Anti-Exploration with Distributional Random Network Distillation
|
https://proceedings.mlr.press/v235/yang24w.html
|
Kai Yang, Jian Tao, Jiafei Lyu, Xiu Li
|
https://proceedings.mlr.press/v235/yang24w.html
|
ICML 2024
|
Exploration remains a critical issue in deep reinforcement learning for an agent to attain high returns in unknown environments. Although the prevailing exploration Random Network Distillation (RND) algorithm has been demonstrated to be effective in numerous environments, it often needs more discriminative power in bonus allocation. This paper highlights the “bonus inconsistency” issue within RND, pinpointing its primary limitation. To address this issue, we introduce the Distributional RND (DRND), a derivative of the RND. DRND enhances the exploration process by distilling a distribution of random networks and implicitly incorporating pseudo counts to improve the precision of bonus allocation. This refinement encourages agents to engage in more extensive exploration. Our method effectively mitigates the inconsistency issue without introducing significant computational overhead. Both theoretical analysis and experimental results demonstrate the superiority of our approach over the original RND algorithm. Our method excels in challenging online exploration scenarios and effectively serves as an anti-exploration mechanism in D4RL offline tasks. Our code is publicly available at https://github.com/yk7333/DRND.
|
https://proceedings.mlr.press/v235/yang24x.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24x/yang24x.pdf
|
https://openreview.net/forum?id=SRmZw7nEGW
|
UniAudio: Towards Universal Audio Generation with Large Language Models
|
https://proceedings.mlr.press/v235/yang24x.html
|
Dongchao Yang, Jinchuan Tian, Xu Tan, Rongjie Huang, Songxiang Liu, Haohan Guo, Xuankai Chang, Jiatong Shi, Sheng Zhao, Jiang Bian, Zhou Zhao, Xixin Wu, Helen M. Meng
|
https://proceedings.mlr.press/v235/yang24x.html
|
ICML 2024
|
Audio generation is a major branch of generative AI research. Compared with prior works in this area that are commonly task-specific with heavy domain knowledge, this paper advocates building universal audio generation models that can handle various tasks in a unified manner. As recent research on large language models (LLMs) has demonstrated their strong ability to handle multiple tasks, this work presents UniAudio, an LLM-based audio generation model that supports a wide range of audio generation tasks. Based on various input conditions, such as phoneme, text description, or audio itself, UniAudio can generate speech, sound, music, and singing voice. The proposed UniAudio is built with 100k hours of multi-source open-available audio data and is scaled to 1B parameters. The audio tokenization method and language model architecture are also specifically designed for both performance and efficiency. Experimentally, UniAuido supports 11 audio generation tasks and achieves competitive results on all tasks consistently. We also show that UniAudio can support new tasks seamlessly via simple fine-tuning.
|
https://proceedings.mlr.press/v235/yang24y.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24y/yang24y.pdf
|
https://openreview.net/forum?id=Uo3LNg5SLY
|
Explain Temporal Black-Box Models via Functional Decomposition
|
https://proceedings.mlr.press/v235/yang24y.html
|
Linxiao Yang, Yunze Tong, Xinyue Gu, Liang Sun
|
https://proceedings.mlr.press/v235/yang24y.html
|
ICML 2024
|
How to explain temporal models is a significant challenge due to the inherent characteristics of time series data, notably the strong temporal dependencies and interactions between observations. Unlike ordinary tabular data, data at different time steps in time series usually interact dynamically, forming influential patterns that shape the model’s predictions, rather than only acting in isolation. Existing explanatory approaches for time series often overlook these crucial temporal interactions by treating time steps as separate entities, leading to a superficial understanding of model behavior. To address this challenge, we introduce FDTempExplainer, an innovative model-agnostic explanation method based on functional decomposition, tailored to unravel the complex interplay within black-box time series models. Our approach disentangles the individual contributions from each time step, as well as the aggregated influence of their interactions, in a rigorous framework. FDTempExplainer accurately measures the strength of interactions, yielding insights that surpass those from baseline models. We demonstrate the effectiveness of our approach in a wide range of time series applications, including anomaly detection, classification, and forecasting, showing its superior performance to the state-of-the-art algorithms.
|
https://proceedings.mlr.press/v235/yang24z.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24z/yang24z.pdf
|
https://openreview.net/forum?id=EZH4CsKV6O
|
Position: Video as the New Language for Real-World Decision Making
|
https://proceedings.mlr.press/v235/yang24z.html
|
Sherry Yang, Jacob C Walker, Jack Parker-Holder, Yilun Du, Jake Bruce, Andre Barreto, Pieter Abbeel, Dale Schuurmans
|
https://proceedings.mlr.press/v235/yang24z.html
|
ICML 2024
|
Both text and video data are abundant on the internet and support large-scale self-supervised learning through next token or frame prediction. However, they have not been equally leveraged: language models have had significant real-world impact, whereas video generation has remained largely limited to media entertainment. Yet video data captures important information about the physical world that is difficult to express in language. To address this gap, we discuss an under-appreciated opportunity to extend video generation to solve tasks in the real world. We observe how, akin to language, video can serve as a unified interface that can absorb internet knowledge and represent diverse tasks. Moreover, we demonstrate how, like language models, video generation can serve as planners, agents, compute engines, and environment simulators through techniques such as in-context learning, planning and reinforcement learning. We identify major impact opportunities in domains such as robotics, self-driving, and science, supported by recent work that demonstrates how such advanced capabilities in video generation are plausibly within reach. Lastly, we identify key challenges in video generation that mitigate progress. Addressing these challenges will enable video generation models to demonstrate unique value alongside language models in a wider array of AI applications.
|
https://proceedings.mlr.press/v235/yang24aa.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24aa/yang24aa.pdf
|
https://openreview.net/forum?id=kIh7GJmRfD
|
ATraDiff: Accelerating Online Reinforcement Learning with Imaginary Trajectories
|
https://proceedings.mlr.press/v235/yang24aa.html
|
Qianlan Yang, Yu-Xiong Wang
|
https://proceedings.mlr.press/v235/yang24aa.html
|
ICML 2024
|
Training autonomous agents with sparse rewards is a long-standing problem in online reinforcement learning (RL), due to low data efficiency. Prior work overcomes this challenge by extracting useful knowledge from offline data, often accomplished through the learning of action distribution from offline data and utilizing the learned distribution to facilitate online RL. However, since the offline data are given and fixed, the extracted knowledge is inherently limited, making it difficult to generalize to new tasks. We propose a novel approach that leverages offline data to learn a generative diffusion model, coined as Adaptive Trajectory Diffuser (ATraDiff). This model generates synthetic trajectories, serving as a form of data augmentation and consequently enhancing the performance of online RL methods. The key strength of our diffuser lies in its adaptability, allowing it to effectively handle varying trajectory lengths and mitigate distribution shifts between online and offline data. Because of its simplicity, ATraDiff seamlessly integrates with a wide spectrum of RL methods. Empirical evaluation shows that ATraDiff consistently achieves state-of-the-art performance across a variety of environments, with particularly pronounced improvements in complicated settings. Our code and demo video are available at https://atradiff.github.io.
|
https://proceedings.mlr.press/v235/yang24ab.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24ab/yang24ab.pdf
|
https://openreview.net/forum?id=ia5XvxFUJT
|
Gated Linear Attention Transformers with Hardware-Efficient Training
|
https://proceedings.mlr.press/v235/yang24ab.html
|
Songlin Yang, Bailin Wang, Yikang Shen, Rameswar Panda, Yoon Kim
|
https://proceedings.mlr.press/v235/yang24ab.html
|
ICML 2024
|
Transformers with linear attention allow for efficient parallel training but can simultaneously be formulated as an RNN with 2D (matrix-valued) hidden states, thus enjoying linear-time inference complexity. However, linear attention generally underperforms ordinary softmax attention. Moreover, current implementations of linear attention lack I/O-awareness and are thus slower than highly optimized implementations of softmax attention. This work describes a hardware-efficient algorithm for linear attention that trades off memory movement against parallelizability. The resulting implementation, dubbed FlashLinearAttention, is faster than FlashAttention-2 as a standalone layer even on short sequence lengths (e.g., 1K). We then generalize this algorithm to a more expressive variant of linear attention with data-dependent gates. When used as a replacement for the standard attention layer in Transformers, the resulting gated linear attention (GLA) Transformer is found to perform competitively against the LLaMA-architecture Transformer as well recent linear-time-inference baselines such as RetNet and Mamba on moderate-scale language modeling experiments. GLA Transformer is especially effective at length generalization, enabling a model trained on 2K to generalize to sequences longer than 20K without significant perplexity degradations. For training speed, the GLA Transformer has higher throughput than a similarly-sized Mamba model.
|
https://proceedings.mlr.press/v235/yang24ac.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24ac/yang24ac.pdf
|
https://openreview.net/forum?id=AQYabSOfci
|
Analysis for Abductive Learning and Neural-Symbolic Reasoning Shortcuts
|
https://proceedings.mlr.press/v235/yang24ac.html
|
Xiao-Wen Yang, Wen-Da Wei, Jie-Jing Shao, Yu-Feng Li, Zhi-Hua Zhou
|
https://proceedings.mlr.press/v235/yang24ac.html
|
ICML 2024
|
Abductive learning models (ABL) and neural-symbolic predictive models (NeSy) have been recently shown effective, as they allow us to infer labels that are consistent with some prior knowledge by reasoning over high-level concepts extracted from sub-symbolic inputs. However, their generalization ability is affected by reasoning shortcuts: high accuracy on given targets but leveraging intermediate concepts with unintended semantics. Although there have been techniques to alleviate reasoning shortcuts, theoretical efforts on this issue remain to be limited. This paper proposes a simple and effective analysis to quantify harm caused by it and how can mitigate it. We quantify three main factors in how NeSy algorithms are affected by reasoning shortcuts: the complexity of the knowledge base, the sample size, and the hypothesis space. In addition, we demonstrate that ABL can reduce shortcut risk by selecting specific distance functions in consistency optimization, thereby demonstrating its potential and approach to solving shortcut problems. Empirical studies demonstrate the rationality of the analysis. Moreover, the proposal is suitable for many ABL and NeSy algorithms and can be easily extended to handle other cases of reasoning shortcuts.
|
https://proceedings.mlr.press/v235/yang24ad.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24ad/yang24ad.pdf
|
https://openreview.net/forum?id=XWkRyIjYDp
|
Stability and Generalization of Stochastic Compositional Gradient Descent Algorithms
|
https://proceedings.mlr.press/v235/yang24ad.html
|
Ming Yang, Xiyuan Wei, Tianbao Yang, Yiming Ying
|
https://proceedings.mlr.press/v235/yang24ad.html
|
ICML 2024
|
Many machine learning tasks can be formulated as a stochastic compositional optimization (SCO) problem such as reinforcement learning, AUC maximization and meta-learning, where the objective function involves a nested composition associated with an expectation. Although many studies have been devoted to studying the convergence behavior of SCO algorithms, there is little work on understanding their generalization, that is, how these learning algorithms built from training data would behave on future test examples. In this paper, we provide the stability and generalization analysis of stochastic compositional gradient descent algorithms in the framework of statistical learning theory. Firstly, we introduce a stability concept called compositional uniform stability and establish its quantitative relation with generalization for SCO problems. Then, we establish the compositional uniform stability results for two notable stochastic compositional gradient descent algorithms, namely SCGD and SCSC. Finally, we derive dimension-independent excess risk bounds for SCGD and SCSC by balancing stability results and optimization errors. To the best of our knowledge, these are the first-ever known results on stability and generalization analysis of stochastic compositional gradient descent algorithms.
|
https://proceedings.mlr.press/v235/yang24ae.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24ae/yang24ae.pdf
|
https://openreview.net/forum?id=Dn4B53IcCW
|
How Graph Neural Networks Learn: Lessons from Training Dynamics
|
https://proceedings.mlr.press/v235/yang24ae.html
|
Chenxiao Yang, Qitian Wu, David Wipf, Ruoyu Sun, Junchi Yan
|
https://proceedings.mlr.press/v235/yang24ae.html
|
ICML 2024
|
A long-standing goal in deep learning has been to characterize the learning behavior of black-box models in a more interpretable manner. For graph neural networks (GNNs), considerable advances have been made in formalizing what functions they can represent, but whether GNNs will learn desired functions during the optimization process remains less clear. To fill this gap, we study their training dynamics in function space. In particular, we find that the optimization of GNNs through gradient descent implicitly leverages the graph structure to update the learned function. This phenomenon is dubbed as kernel-graph alignment, which has been empirically and theoretically corroborated. This new analytical framework from the optimization perspective enables interpretable explanations of when and why the learned GNN functions generalize, which are relevant to their limitations on heterophilic graphs. From a practical standpoint, it also provides high-level principles for designing new algorithms. We exemplify this by showing that a simple and efficient non-parametric algorithm, obtained by explicitly using graph structure to update the learned function, can consistently compete with nonlinear GNNs.
|
https://proceedings.mlr.press/v235/yang24af.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24af/yang24af.pdf
|
https://openreview.net/forum?id=ebt5BfRHcW
|
Harnessing Hierarchical Label Distribution Variations in Test Agnostic Long-tail Recognition
|
https://proceedings.mlr.press/v235/yang24af.html
|
Zhiyong Yang, Qianqian Xu, Zitai Wang, Sicong Li, Boyu Han, Shilong Bao, Xiaochun Cao, Qingming Huang
|
https://proceedings.mlr.press/v235/yang24af.html
|
ICML 2024
|
This paper explores test-agnostic long-tail recognition, a challenging long-tail task where the test label distributions are unknown and arbitrarily imbalanced. We argue that the variation in these distributions can be broken down hierarchically into global and local levels. The global ones reflect a broad range of diversity, while the local ones typically arise from milder changes, often focused On a particular neighbor. Traditional methods predominantly use a Mixture-of-Expert (MoE) approach, targeting a few fixed test label distributions that exhibit substantial global variations. However, the local variations are left unconsidered. To address this issue, we propose a new MoE strategy, $\mathsf{DirMixE}$, which assigns experts to different Dirichlet meta-distributions of the label distribution, each targeting a specific aspect of local variations. Additionally, the diversity among these Dirichlet meta-distributions inherently captures global variations. This dual-level approach also leads to a more stable objective function, allowing us to sample different test distributions better to quantify the mean and variance of performance outcomes. Theoretically, we show that our proposed objective benefits from enhanced generalization by virtue of the variance-based regularization. Comprehensive experiments across multiple benchmarks confirm the effectiveness of $\mathsf{DirMixE}$.
|
https://proceedings.mlr.press/v235/yang24ag.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24ag/yang24ag.pdf
|
https://openreview.net/forum?id=HDrXBr26UI
|
Neuro-Symbolic Temporal Point Processes
|
https://proceedings.mlr.press/v235/yang24ag.html
|
Yang Yang, Chao Yang, Boyang Li, Yinghao Fu, Shuang Li
|
https://proceedings.mlr.press/v235/yang24ag.html
|
ICML 2024
|
Our goal is to $\textit{efficiently}$ discover a compact set of temporal logic rules to explain irregular events of interest. We introduce a neural-symbolic rule induction framework within the temporal point process model. The negative log-likelihood is the loss that guides the learning, where the explanatory logic rules and their weights are learned end-to-end in a $\textit{differentiable}$ way. Specifically, predicates and logic rules are represented as $\textit{vector embeddings}$, where the predicate embeddings are fixed and the rule embeddings are trained via gradient descent to obtain the most appropriate compositional representations of the predicate embeddings. To make the rule learning process more efficient and flexible, we adopt a $\textit{sequential covering algorithm}$, which progressively adds rules to the model and removes the event sequences that have been explained until all event sequences have been covered. All the found rules will be fed back to the models for a final rule embedding and weight refinement. Our approach showcases notable efficiency and accuracy across synthetic and real datasets, surpassing state-of-the-art baselines by a wide margin in terms of efficiency.
|
https://proceedings.mlr.press/v235/yang24ah.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24ah/yang24ah.pdf
|
https://openreview.net/forum?id=01ahsMovBx
|
One Meta-tuned Transformer is What You Need for Few-shot Learning
|
https://proceedings.mlr.press/v235/yang24ah.html
|
Xu Yang, Huaxiu Yao, Ying Wei
|
https://proceedings.mlr.press/v235/yang24ah.html
|
ICML 2024
|
Pre-trained vision transformers have revolutionized few-shot image classification, and it has been recently demonstrated that the previous common practice of meta-learning in synergy with these pre-trained transformers still holds significance. In this work, we design a new framework centered exclusively on self-attention, called MetaFormer, which extends the vision transformers beyond patch token interactions to encompass relationships between samples and tasks simultaneously for further advancing their downstream task performance. Leveraging the intrinsical property of ViTs in handling local patch relationships, we propose Masked Sample Attention (MSA) to efficiently embed the sample relationships into the network, where an adaptive mask is attached for enhancing task-specific feature consistency and providing flexibility in switching between few-shot learning setups. To encapsulate task relationships while filtering out background noise, Patch-grained Task Attention (PTA) is designed to maintain a dynamic knowledge pool consolidating diverse patterns from historical tasks. MetaFormer demonstrates coherence and compatibility with off-the-shelf pre-trained vision transformers and shows significant improvements in both inductive and transductive few-shot learning scenarios, outperforming state-of-the-art methods by up to 8.77% and 6.25% on 12 in-domain and 10 cross-domain datasets, respectively.
|
https://proceedings.mlr.press/v235/yang24ai.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24ai/yang24ai.pdf
|
https://openreview.net/forum?id=DgLFkAPwuZ
|
Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs
|
https://proceedings.mlr.press/v235/yang24ai.html
|
Ling Yang, Zhaochen Yu, Chenlin Meng, Minkai Xu, Stefano Ermon, Bin Cui
|
https://proceedings.mlr.press/v235/yang24ai.html
|
ICML 2024
|
Diffusion models have exhibit exceptional performance in text-to-image generation and editing. However, existing methods often face challenges when handling complex text prompts that involve multiple objects with multiple attributes and relationships. In this paper, we propose a brand new training-free text-to-image generation/editing framework, namely Recaption, Plan and Generate (RPG), harnessing the powerful chain-of-thought reasoning ability of multimodal LLMs to enhance the compositionality of text-to-image diffusion models. Our approach employs the MLLM as a global planner to decompose the process of generating complex images into multiple simpler generation tasks within subregions. We propose complementary regional diffusion to enable region-wise compositional generation. Furthermore, we integrate text-guided image generation and editing within the proposed RPG in a closed-loop fashion, thereby enhancing generalization ability. Extensive experiments demonstrate our RPG outperforms state-of-the-art text-to-image models, including DALL-E 3 and SDXL, particularly in multi-category object composition and text-image semantic alignment. Notably, our RPG framework exhibits wide compatibility with various MLLM architectures and diffusion backbones. Our code is available at https://github.com/YangLing0818/RPG-DiffusionMaster
|
https://proceedings.mlr.press/v235/yang24aj.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24aj/yang24aj.pdf
|
https://openreview.net/forum?id=46vXhZn7lN
|
Accelerating Look-ahead in Bayesian Optimization: Multilevel Monte Carlo is All you Need
|
https://proceedings.mlr.press/v235/yang24aj.html
|
Shangda Yang, Vitaly Zankin, Maximilian Balandat, Stefan Scherer, Kevin Thomas Carlberg, Neil Walton, Kody J. H. Law
|
https://proceedings.mlr.press/v235/yang24aj.html
|
ICML 2024
|
We leverage multilevel Monte Carlo (MLMC) to improve the performance of multi-step look- ahead Bayesian optimization (BO) methods that involve nested expectations and maximizations. Often these expectations must be computed by Monte Carlo (MC). The complexity rate of naive MC degrades for nested operations, whereas MLMC is capable of achieving the canonical MC convergence rate for this type of problem, independently of dimension and without any smoothness assumptions. Our theoretical study focuses on the approximation improvements for two- and three-step look-ahead acquisition functions, but, as we discuss, the approach is generalizable in various ways, including beyond the context of BO. Our findings are verified numerically and the benefits of MLMC for BO are illustrated on several benchmark examples. Code is available at https://github.com/Shangda-Yang/MLMCBO.
|
https://proceedings.mlr.press/v235/yang24ak.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24ak/yang24ak.pdf
|
https://openreview.net/forum?id=ZTN866OsGx
|
MorphGrower: A Synchronized Layer-by-layer Growing Approach for Plausible Neuronal Morphology Generation
|
https://proceedings.mlr.press/v235/yang24ak.html
|
Nianzu Yang, Kaipeng Zeng, Haotian Lu, Yexin Wu, Zexin Yuan, Danni Chen, Shengdian Jiang, Jiaxiang Wu, Yimin Wang, Junchi Yan
|
https://proceedings.mlr.press/v235/yang24ak.html
|
ICML 2024
|
Neuronal morphology is essential for studying brain functioning and understanding neurodegenerative disorders. As acquiring real-world morphology data is expensive, computational approaches for morphology generation have been studied. Traditional methods heavily rely on expert-set rules and parameter tuning, making it difficult to generalize across different types of morphologies. Recently, MorphVAE was introduced as the sole learning-based method, but its generated morphologies lack plausibility, i.e., they do not appear realistic enough and most of the generated samples are topologically invalid. To fill this gap, this paper proposes MorphGrower, which mimicks the neuron natural growth mechanism for generation. Specifically, MorphGrower generates morphologies layer by layer, with each subsequent layer conditioned on the previously generated structure. During each layer generation, MorphGrower utilizes a pair of sibling branches as the basic generation block and generates branch pairs synchronously. This approach ensures topological validity and allows for fine-grained generation, thereby enhancing the realism of the final generated morphologies. Results on four real-world datasets demonstrate that MorphGrower outperforms MorphVAE by a notable margin. Importantly, the electrophysiological response simulation demonstrates the plausibility of our generated samples from a neuroscience perspective. Our code is available at https://github.com/Thinklab-SJTU/MorphGrower.
|
https://proceedings.mlr.press/v235/yang24al.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24al/yang24al.pdf
|
https://openreview.net/forum?id=inEuvSg0y1
|
Mol-AE: Auto-Encoder Based Molecular Representation Learning With 3D Cloze Test Objective
|
https://proceedings.mlr.press/v235/yang24al.html
|
Junwei Yang, Kangjie Zheng, Siyu Long, Zaiqing Nie, Ming Zhang, Xinyu Dai, Wei-Ying Ma, Hao Zhou
|
https://proceedings.mlr.press/v235/yang24al.html
|
ICML 2024
|
3D molecular representation learning has gained tremendous interest and achieved promising performance in various downstream tasks. A series of recent approaches follow a prevalent framework: an encoder-only model coupled with a coordinate denoising objective. However, through a series of analytical experiments, we prove that the encoder-only model with coordinate denoising objective exhibits inconsistency between pre-training and downstream objectives, as well as issues with disrupted atomic identifiers. To address these two issues, we propose Mol-AE for molecular representation learning, an auto-encoder model using positional encoding as atomic identifiers. We also propose a new training objective named 3D Cloze Test to make the model learn better atom spatial relationships from real molecular substructures. Empirical results demonstrate that Mol-AE achieves a large margin performance gain compared to the current state-of-the-art 3D molecular modeling approach.
|
https://proceedings.mlr.press/v235/yang24am.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yang24am/yang24am.pdf
|
https://openreview.net/forum?id=z8sYc334fU
|
What is Dataset Distillation Learning?
|
https://proceedings.mlr.press/v235/yang24am.html
|
William Yang, Ye Zhu, Zhiwei Deng, Olga Russakovsky
|
https://proceedings.mlr.press/v235/yang24am.html
|
ICML 2024
|
Dataset distillation has emerged as a strategy to overcome the hurdles associated with large datasets by learning a compact set of synthetic data that retains essential information from the original dataset. While distilled data can be used to train high performing models, little is understood about how the information is stored. In this study, we posit and answer three questions about the behavior, representativeness, and point-wise information content of distilled data. We reveal distilled data cannot serve as a substitute for real data during training outside the standard evaluation setting for dataset distillation. Additionally, the distillation process retains high task performance by compressing information related to the early training dynamics of real models. Finally, we provide an framework for interpreting distilled data and reveal that individual distilled data points contain meaningful semantic information. This investigation sheds light on the intricate nature of distilled data, providing a better understanding on how they can be effectively utilized.
|
https://proceedings.mlr.press/v235/wang24cv.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/wang24cv/wang24cv.pdf
|
https://openreview.net/forum?id=aC1LSa4nXs
|
Protein Conformation Generation via Force-Guided SE(3) Diffusion Models
|
https://proceedings.mlr.press/v235/wang24cv.html
|
Yan Wang, Lihao Wang, Yuning Shen, Yiqun Wang, Huizhuo Yuan, Yue Wu, Quanquan Gu
|
https://proceedings.mlr.press/v235/wang24cv.html
|
ICML 2024
|
The conformational landscape of proteins is crucial to understanding their functionality in complex biological processes. Traditional physics-based computational methods, such as molecular dynamics (MD) simulations, suffer from rare event sampling and long equilibration time problems, hindering their applications in general protein systems. Recently, deep generative modeling techniques, especially diffusion models, have been employed to generate novel protein conformations. However, existing score-based diffusion methods cannot properly incorporate important physical prior knowledge to guide the generation process, causing large deviations in the sampled protein conformations from the equilibrium distribution. In this paper, to overcome these limitations, we propose a force-guided $\mathrm{SE}(3)$ diffusion model, ConfDiff, for protein conformation generation. By incorporating a force-guided network with a mixture of data-based score models, ConfDiff can generate protein conformations with rich diversity while preserving high fidelity. Experiments on a variety of protein conformation prediction tasks, including 12 fast-folding proteins and the Bovine Pancreatic Trypsin Inhibitor (BPTI), demonstrate that our method surpasses the state-of-the-art method.
|
https://proceedings.mlr.press/v235/yao24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yao24a/yao24a.pdf
|
https://openreview.net/forum?id=u9oSQtujCF
|
Empowering Graph Invariance Learning with Deep Spurious Infomax
|
https://proceedings.mlr.press/v235/yao24a.html
|
Tianjun Yao, Yongqiang Chen, Zhenhao Chen, Kai Hu, Zhiqiang Shen, Kun Zhang
|
https://proceedings.mlr.press/v235/yao24a.html
|
ICML 2024
|
Recently, there has been a surge of interest in developing graph neural networks that utilize the invariance principle on graphs to generalize the out-of-distribution (OOD) data. Due to the limited knowledge about OOD data, existing approaches often pose assumptions about the correlation strengths of the underlying spurious features and the target labels. However, this prior is often unavailable and will change arbitrarily in the real-world scenarios, which may lead to severe failures of the existing graph invariance learning methods. To bridge this gap, we introduce a novel graph invariance learning paradigm, which induces a robust and general inductive bias, which is built upon the observation that the infomax principle encourages learning spurious features regardless of spurious correlation strengths. We further propose the EQuAD framework that realizes this learning paradigm and employs tailored learning objectives that provably elicit invariant features by disentangling them from the spurious features learned through infomax. Notably, EQuAD shows stable and enhanced performance across different degrees of bias in synthetic datasets and challenging real-world datasets up to 31.76%.
|
https://proceedings.mlr.press/v235/yao24b.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yao24b/yao24b.pdf
|
https://openreview.net/forum?id=dT6ZbSxh33
|
Human vs. Generative AI in Content Creation Competition: Symbiosis or Conflict?
|
https://proceedings.mlr.press/v235/yao24b.html
|
Fan Yao, Chuanhao Li, Denis Nekipelov, Hongning Wang, Haifeng Xu
|
https://proceedings.mlr.press/v235/yao24b.html
|
ICML 2024
|
The advent of generative AI (GenAI) technology produces a transformative impact on the content creation landscape, offering alternative approaches to produce diverse, good-quality content across media, thereby reshaping online ecosystems but also raising concerns about market over-saturation and the potential marginalization of human creativity. Our work introduces a competition model generalized from the Tullock contest to analyze the tension between human creators and GenAI. Our theory and simulations suggest that despite challenges, a stable equilibrium between human and AI-generated content is possible. Our work contributes to understanding the competitive dynamics in the content creation industry, offering insights into the future interplay between human creativity and technological advancements in GenAI.
|
https://proceedings.mlr.press/v235/yao24c.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yao24c/yao24c.pdf
|
https://openreview.net/forum?id=VHtIDVaOKC
|
Mobile Attention: Mobile-Friendly Linear-Attention for Vision Transformers
|
https://proceedings.mlr.press/v235/yao24c.html
|
Zhiyu Yao, Jian Wang, Haixu Wu, Jingdong Wang, Mingsheng Long
|
https://proceedings.mlr.press/v235/yao24c.html
|
ICML 2024
|
Vision Transformers (ViTs) excel in computer vision tasks due to their ability to capture global context among tokens. However, their quadratic complexity $\mathcal{O}(N^2D)$ in terms of token number $N$ and feature dimension $D$ limits practical use on mobile devices, necessitating more mobile-friendly ViTs with reduced latency. Multi-head linear-attention is emerging as a promising alternative with linear complexity $\mathcal{O}(NDd)$, where $d$ is the per-head dimension. Still, more compute is needed as $d$ gets large for model accuracy. Reducing $d$ improves mobile friendliness at the expense of excessive small heads weak at learning valuable subspaces, ultimately impeding model capability. To overcome this efficiency-capability dilemma, we propose a novel Mobile-Attention design with a head-competition mechanism empowered by information flow, which prevents overemphasis on less important subspaces upon trivial heads while preserving essential subspaces to ensure Transformer’s capability. It enables linear-time complexity on mobile devices by supporting a small per-head dimension $d$ for mobile efficiency. By replacing the standard attention of ViTs with Mobile-Attention, our optimized ViTs achieved enhanced model capacity and competitive performance in a range of computer vision tasks. Specifically, we have achieved remarkable reductions in latency on the iPhone 12. Code is available at https://github.com/thuml/MobileAttention.
|
https://proceedings.mlr.press/v235/yao24d.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yao24d/yao24d.pdf
|
https://openreview.net/forum?id=aaeJpJw5Ur
|
Socialized Learning: Making Each Other Better Through Multi-Agent Collaboration
|
https://proceedings.mlr.press/v235/yao24d.html
|
Xinjie Yao, Yu Wang, Pengfei Zhu, Wanyu Lin, Jialu Li, Weihao Li, Qinghua Hu
|
https://proceedings.mlr.press/v235/yao24d.html
|
ICML 2024
|
Learning new knowledge frequently occurs in our dynamically changing world, e.g., humans culturally evolve by continuously acquiring new abilities to sustain their survival, leveraging collective intelligence rather than a large number of individual attempts. The effective learning paradigm during cultural evolution is termed socialized learning (SL). Consequently, a straightforward question arises: Can multi-agent systems acquire more new abilities like humans? In contrast to most existing methods that address continual learning and multi-agent collaboration, our emphasis lies in a more challenging problem: we prioritize the knowledge in the original expert classes, and as we adeptly learn new ones, the accuracy in the original expert classes stays superior among all in a directional manner. Inspired by population genetics and cognitive science, leading to unique and complete development, we propose Multi-Agent Socialized Collaboration (MASC), which achieves SL through interactions among multiple agents. Specifically, we introduce collective collaboration and reciprocal altruism modules, organizing collaborative behaviors, promoting information sharing, and facilitating learning and knowledge interaction among individuals. We demonstrate the effectiveness of multi-agent collaboration in an extensive empirical study. Our code will be publicly available at https://github.com/yxjdarren/SL.
|
https://proceedings.mlr.press/v235/yaras24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yaras24a/yaras24a.pdf
|
https://openreview.net/forum?id=uDkXoZMzBv
|
Compressible Dynamics in Deep Overparameterized Low-Rank Learning & Adaptation
|
https://proceedings.mlr.press/v235/yaras24a.html
|
Can Yaras, Peng Wang, Laura Balzano, Qing Qu
|
https://proceedings.mlr.press/v235/yaras24a.html
|
ICML 2024
|
While overparameterization in machine learning models offers great benefits in terms of optimization and generalization, it also leads to increased computational requirements as model sizes grow. In this work, we show that by leveraging the inherent low-dimensional structures of data and compressible dynamics within the model parameters, we can reap the benefits of overparameterization without the computational burdens. In practice, we demonstrate the effectiveness of this approach for deep low-rank matrix completion as well as fine-tuning language models. Our approach is grounded in theoretical findings for deep overparameterized low-rank matrix recovery, where we show that the learning dynamics of each weight matrix are confined to an invariant low-dimensional subspace. Consequently, we can construct and train compact, highly compressed factorizations possessing the same benefits as their overparameterized counterparts. In the context of deep matrix completion, our technique substantially improves training efficiency while retaining the advantages of overparameterization. For language model fine-tuning, we propose a method called "Deep LoRA", which improves the existing low-rank adaptation (LoRA) technique, leading to reduced overfitting and a simplified hyperparameter setup, while maintaining comparable efficiency. We validate the effectiveness of Deep LoRA on natural language tasks, particularly when fine-tuning with limited data.
|
https://proceedings.mlr.press/v235/yau24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yau24a/yau24a.pdf
|
https://openreview.net/forum?id=GxOFM3f5Vm
|
EMC$^2$: Efficient MCMC Negative Sampling for Contrastive Learning with Global Convergence
|
https://proceedings.mlr.press/v235/yau24a.html
|
Chung-Yiu Yau, Hoi To Wai, Parameswaran Raman, Soumajyoti Sarkar, Mingyi Hong
|
https://proceedings.mlr.press/v235/yau24a.html
|
ICML 2024
|
A key challenge in contrastive learning is to generate negative samples from a large sample set to contrast with positive samples, for learning better encoding of the data. These negative samples often follow a softmax distribution which are dynamically updated during the training process. However, sampling from this distribution is non-trivial due to the high computational costs in computing the partition function. In this paper, we propose an $\underline{\text{E}}$fficient $\underline{\text{M}}$arkov $\underline{\text{C}}$hain Monte Carlo negative sampling method for $\underline{\text{C}}$ontrastive learning (EMC$^2$). We follow the global contrastive learning loss as introduced in SogCLR, and propose EMC$^2$ which utilizes an adaptive Metropolis-Hastings subroutine to generate hardness-aware negative samples in an online fashion during the optimization. We prove that EMC$^2$ finds an $\mathcal{O}(1/\sqrt{T})$-stationary point of the global contrastive loss in $T$ iterations. Compared to prior works, EMC$^2$ is the first algorithm that exhibits global convergence (to stationarity) regardless of the choice of batch size while exhibiting low computation and memory cost. Numerical experiments validate that EMC$^2$ is effective with small batch training and achieves comparable or better performance than baseline algorithms. We report the results for pre-training image encoders on STL-10 and Imagenet-100.
|
https://proceedings.mlr.press/v235/ye24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/ye24a/ye24a.pdf
|
https://openreview.net/forum?id=Z0S6fUdW68
|
Towards Robust Model-Based Reinforcement Learning Against Adversarial Corruption
|
https://proceedings.mlr.press/v235/ye24a.html
|
Chenlu Ye, Jiafan He, Quanquan Gu, Tong Zhang
|
https://proceedings.mlr.press/v235/ye24a.html
|
ICML 2024
|
This study tackles the challenges of adversarial corruption in model-based reinforcement learning (RL), where the transition dynamics can be corrupted by an adversary. Existing studies on corruption-robust RL mostly focus on the setting of model-free RL, where robust least-square regression is often employed for value function estimation. However, these techniques cannot be directly applied to model-based RL. In this paper, we focus on model-based RL and take the maximum likelihood estimation (MLE) approach to learn transition model. Our work encompasses both online and offline settings. In the online setting, we introduce an algorithm called corruption-robust optimistic MLE (CR-OMLE), which leverages total-variation (TV)-based information ratios as uncertainty weights for MLE. We prove that CR-OMLE achieves a regret of $\tilde{\mathcal{O}}(\sqrt{T} + C)$, where $C$ denotes the cumulative corruption level after $T$ episodes. We also prove a lower bound to show that the additive dependence on $C$ is optimal. We extend our weighting technique to the offline setting, and propose an algorithm named corruption-robust pessimistic MLE (CR-PMLE). Under a uniform coverage condition, CR-PMLE exhibits suboptimality worsened by $\mathcal{O}(C/n)$, nearly matching the lower bound. To the best of our knowledge, this is the first work on corruption-robust model-based RL algorithms with provable guarantees.
|
https://proceedings.mlr.press/v235/yen24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yen24a/yen24a.pdf
|
https://openreview.net/forum?id=h2uBuQvpp8
|
Adaptive Sampling of k-Space in Magnetic Resonance for Rapid Pathology Prediction
|
https://proceedings.mlr.press/v235/yen24a.html
|
Chen-Yu Yen, Raghav Singhal, Umang Sharma, Rajesh Ranganath, Sumit Chopra, Lerrel Pinto
|
https://proceedings.mlr.press/v235/yen24a.html
|
ICML 2024
|
Magnetic Resonance (MR) imaging, despite its proven diagnostic utility, remains an inaccessible imaging modality for disease surveillance at the population level. A major factor rendering MR inaccessible is lengthy scan times. An MR scanner collects measurements associated with the underlying anatomy in the Fourier space, also known as the k-space. Creating a high-fidelity image requires collecting large quantities of such measurements, increasing the scan time. Traditionally to accelerate an MR scan, image reconstruction from under-sampled k-space data is the method of choice. However, recent works show the feasibility of bypassing image reconstruction and directly learning to detect disease directly from a sparser learned subset of the k-space measurements. In this work, we propose Adaptive Sampling for MR (ASMR), a sampling method that learns an adaptive policy to sequentially select k-space samples to optimize for target disease detection. On 6 out of 8 pathology classification tasks spanning the Knee, Brain, and Prostate MR scans, ASMR reaches within 2% of the performance of a fully sampled classifier while using only 8% of the k-space, as well as outperforming prior state-of-the-art work in k-space sampling such as EMRT, LOUPE, and DPS.
|
https://proceedings.mlr.press/v235/yin24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yin24a/yin24a.pdf
|
https://openreview.net/forum?id=GFfWzAReAc
|
StableMask: Refining Causal Masking in Decoder-only Transformer
|
https://proceedings.mlr.press/v235/yin24a.html
|
Qingyu Yin, Xuzheng He, Xiang Zhuang, Yu Zhao, Jianhua Yao, Xiaoyu Shen, Qiang Zhang
|
https://proceedings.mlr.press/v235/yin24a.html
|
ICML 2024
|
The decoder-only Transformer architecture with causal masking and relative position encoding (RPE) has become the de facto choice in language modeling. Despite its exceptional performance across various tasks, we have identified two limitations: First, it prevents all attended tokens from having zero weights during the softmax stage, even if the current embedding has sufficient self-contained information. This compels the model to assign disproportional excessive attention to specific tokens. Second, RPE-based Transformers are not universal approximators due to their limited capacity at encoding absolute positional information, which limits their application in position-critical tasks. In this work, we propose StableMask: a parameter-free method to address both limitations by refining the causal mask. It introduces pseudo-attention values to balance attention distributions and encodes absolute positional information via a progressively decreasing mask ratio. StableMask’s effectiveness is validated both theoretically and empirically, showing significant enhancements in language models with parameter sizes ranging from 71M to 1.4B across diverse datasets and encoding methods. We further show that it supports integration with existing optimization techniques, making it easily usable in practical applications.
|
https://proceedings.mlr.press/v235/yin24b.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yin24b/yin24b.pdf
|
https://openreview.net/forum?id=EfUrTeuUfy
|
Junk DNA Hypothesis: Pruning Small Pre-Trained Weights $\textitIrreversibly$ and $\textitMonotonically$ Impairs “Difficult" Downstream Tasks in LLMs
|
https://proceedings.mlr.press/v235/yin24b.html
|
Lu Yin, Ajay Kumar Jaiswal, Shiwei Liu, Souvik Kundu, Zhangyang Wang
|
https://proceedings.mlr.press/v235/yin24b.html
|
ICML 2024
|
We present Junk DNA Hypothesis by adopting a novel task-centric angle for the pre-trained weights of large language models (LLMs). It has been believed that weights in LLMs contain significant redundancy, leading to the conception that a considerable chunk of the parameters can be removed by pruning without compromising performance. Contrary to this belief, this paper presents a counter-argument: small-magnitude weights of pre-trained model weights encode vital knowledge essential for tackling difficult downstream tasks - manifested as the monotonic relationship between the performance drop of downstream tasks across the difficulty spectrum, as we prune more pre-trained weights by magnitude. Moreover, we reveal that these seemingly inconsequential weights can result in irreparable loss of knowledge and performance degradation in difficult tasks, even when downstream continual training is allowed. Interestingly, our evaluations show that the other popular compression, namely quantization fail to exhibit similar “monotonic" effect and does not as convincingly disentangle this task-difficulty information. To study formally, we introduce several quantifiable metrics to gauge the downstream task difficulty: (a) within the same task category, and (b) across different task categories. Our extensive experiments substantiate the Junk DNA Hypothesis across a diverse range of model sizes, tasks, datasets, and even pruning methods. Codes are available at https://github.com/VITA-Group/Junk_DNA_Hypothesis.git.
|
https://proceedings.mlr.press/v235/yin24c.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yin24c/yin24c.pdf
|
https://openreview.net/forum?id=7DbIyQlfaO
|
Characterizing Truthfulness in Large Language Model Generations with Local Intrinsic Dimension
|
https://proceedings.mlr.press/v235/yin24c.html
|
Fan Yin, Jayanth Srinivasa, Kai-Wei Chang
|
https://proceedings.mlr.press/v235/yin24c.html
|
ICML 2024
|
We study how to characterize and predict the truthfulness of texts generated from large language models (LLMs), which serves as a crucial step in building trust between humans and LLMs. Although several approaches based on entropy or verbalized uncertainty have been proposed to calibrate model predictions, these methods are often intractable, sensitive to hyperparameters, and less reliable when applied in generative tasks with LLMs. In this paper, we suggest investigating internal activations and quantifying LLM’s truthfulness using the local intrinsic dimension (LID) of model activations. Through experiments on four question answering (QA) datasets, we demonstrate the effectiveness of our proposed method. Additionally, we study intrinsic dimensions in LLMs and their relations with model layers, autoregressive language modeling, and the training of LLMs, revealing that intrinsic dimensions can be a powerful approach to understanding LLMs.
|
https://proceedings.mlr.press/v235/yin24d.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yin24d/yin24d.pdf
|
https://openreview.net/forum?id=beXQVQorse
|
High-Dimensional Bayesian Optimization via Semi-Supervised Learning with Optimized Unlabeled Data Sampling
|
https://proceedings.mlr.press/v235/yin24d.html
|
Yuxuan Yin, Yu Wang, Peng Li
|
https://proceedings.mlr.press/v235/yin24d.html
|
ICML 2024
|
We introduce a novel semi-supervised learning approach, named Teacher-Student Bayesian Optimization ($\texttt{TSBO}$), integrating the teacher-student paradigm into BO to minimize expensive labeled data queries for the first time. $\texttt{TSBO}$ incorporates a teacher model, an unlabeled data sampler, and a student model. The student is trained on unlabeled data locations generated by the sampler, with pseudo labels predicted by the teacher. The interplay between these three components implements a unique selective regularization to the teacher in the form of student feedback. This scheme enables the teacher to predict high-quality pseudo labels, enhancing the generalization of the GP surrogate model in the search space. To fully exploit $\texttt{TSBO}$, we propose two optimized unlabeled data samplers to construct effective student feedback that well aligns with the objective of Bayesian optimization. Furthermore, we quantify and leverage the uncertainty of the teacher-student model for the provision of reliable feedback to the teacher in the presence of risky pseudo-label predictions. $\texttt{TSBO}$ demonstrates significantly improved sample-efficiency in several global optimization tasks under tight labeled data budgets. The implementation is available at https://github.com/reminiscenty/TSBO-Official.
|
https://proceedings.mlr.press/v235/yin24e.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yin24e/yin24e.pdf
|
https://openreview.net/forum?id=ahEm3l2P6w
|
Outlier Weighed Layerwise Sparsity (OWL): A Missing Secret Sauce for Pruning LLMs to High Sparsity
|
https://proceedings.mlr.press/v235/yin24e.html
|
Lu Yin, You Wu, Zhenyu Zhang, Cheng-Yu Hsieh, Yaqing Wang, Yiling Jia, Gen Li, Ajay Kumar Jaiswal, Mykola Pechenizkiy, Yi Liang, Michael Bendersky, Zhangyang Wang, Shiwei Liu
|
https://proceedings.mlr.press/v235/yin24e.html
|
ICML 2024
|
Large Language Models (LLMs), renowned for their remarkable performance across diverse domains, present a challenge due to their colossal model size when it comes to practical deployment. In response to this challenge, efforts have been directed toward the application of traditional network pruning techniques to LLMs, uncovering a massive number of parameters can be pruned in one-shot without hurting performance. Building upon insights gained from pre-LLM models, particularly BERT-level language models, prevailing LLM pruning strategies have consistently adhered to the practice of uniformly pruning all layers at equivalent sparsity levels, resulting in robust performance. However, this observation stands in contrast to the prevailing trends observed in the field of vision models, where non-uniform layerwise sparsity typically yields substantially improved results. To elucidate the underlying reasons for this disparity, we conduct a comprehensive analysis of the distribution of token features within LLMs. In doing so, we discover a strong correlation with the emergence of outliers, defined as features exhibiting significantly greater magnitudes compared to their counterparts in feature dimensions. Inspired by this finding, we introduce a novel LLM pruning methodology that incorporates a tailored set of non-uniform layerwise sparsity ratios specifically designed for LLM pruning, termed as Outlier Weighed Layerwise sparsity (OWL). The sparsity ratio of OWL is directly proportional to the outlier ratio observed within each layer, facilitating a more effective alignment between layerwise weight sparsity and outlier ratios. Our empirical evaluation, conducted across the LLaMA-V1/V2, Vicuna, OPT, and Mistral, spanning various benchmarks, demonstrates the distinct advantages offered by OWL over previous methods. For instance, OWL exhibits a remarkable performance gain, surpassing the state-of-the-art Wanda and SparseGPT by 61.22 and 6.80 perplexity at a high sparsity level of 70%, respectively, while delivering 2.6$\times$ end-to-end inference speed-up in the DeepSparse inference engine. Code is available at https://github.com/luuyin/OWL.git.
|
https://proceedings.mlr.press/v235/ying24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/ying24a/ying24a.pdf
|
https://openreview.net/forum?id=R4Ng8zYaiz
|
MMT-Bench: A Comprehensive Multimodal Benchmark for Evaluating Large Vision-Language Models Towards Multitask AGI
|
https://proceedings.mlr.press/v235/ying24a.html
|
Kaining Ying, Fanqing Meng, Jin Wang, Zhiqian Li, Han Lin, Yue Yang, Hao Zhang, Wenbo Zhang, Yuqi Lin, Shuo Liu, Jiayi Lei, Quanfeng Lu, Runjian Chen, Peng Xu, Renrui Zhang, Haozhe Zhang, Peng Gao, Yali Wang, Yu Qiao, Ping Luo, Kaipeng Zhang, Wenqi Shao
|
https://proceedings.mlr.press/v235/ying24a.html
|
ICML 2024
|
Large Vision-Language Models (LVLMs) show significant strides in general-propose multimodal applications such as visual dialogue and embodied navigation. However, existing multimodal evaluation benchmarks cover a limited number of multimodal tasks testing rudimentary capabilities, falling short in tracking LVLM development. In this study, we present MMT-Bench, a comprehensive benchmark designed to assess LVLMs across massive multimodal tasks requiring expert knowledge and deliberate visual recognition, localization, and reasoning. MMT-Bench comprises $31,325$ meticulously curated multi-choice visual questions from various multimodal scenarios such as vehicle driving and embodied navigation, covering $32$ core meta-tasks and $162$ subtasks in multimodal understanding. Due to its extensive task coverage, MMT-Bench enables the evaluation of LVLMs using a task map, facilitating the discovery of in- and out-of-domain tasks. Evaluation results involving $20$ publicly available LVLMs such as the proprietary GeminiProVision model, underscore the significant challenges posed by MMT-Bench. We anticipate that MMT-Bench will inspire the community to develop next-generation multimodal foundation models aimed at achieving general-purpose multimodal intelligence.
|
https://proceedings.mlr.press/v235/yoo24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yoo24a/yoo24a.pdf
|
https://openreview.net/forum?id=Lg8nw3ltvX
|
Layerwise Proximal Replay: A Proximal Point Method for Online Continual Learning
|
https://proceedings.mlr.press/v235/yoo24a.html
|
Jinsoo Yoo, Yunpeng Liu, Frank Wood, Geoff Pleiss
|
https://proceedings.mlr.press/v235/yoo24a.html
|
ICML 2024
|
In online continual learning, a neural network incrementally learns from a non-i.i.d. data stream. Nearly all online continual learning methods employ experience replay to simultaneously prevent catastrophic forgetting and underfitting on past data. Our work demonstrates a limitation of this approach: neural networks trained with experience replay tend to have unstable optimization trajectories, impeding their overall accuracy. Surprisingly, these instabilities persist even when the replay buffer stores all previous training examples, suggesting that this issue is orthogonal to catastrophic forgetting. We minimize these instabilities through a simple modification of the optimization geometry. Our solution, Layerwise Proximal Replay (LPR), balances learning from new and replay data while only allowing for gradual changes in the hidden activation of past data. We demonstrate that LPR consistently improves replay-based online continual learning across multiple problem settings, regardless of the amount of available replay memory.
|
https://proceedings.mlr.press/v235/yoon24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yoon24a/yoon24a.pdf
|
https://openreview.net/forum?id=JtkruFHcRK
|
Uncertainty Estimation by Density Aware Evidential Deep Learning
|
https://proceedings.mlr.press/v235/yoon24a.html
|
Taeseong Yoon, Heeyoung Kim
|
https://proceedings.mlr.press/v235/yoon24a.html
|
ICML 2024
|
Evidential deep learning (EDL) has shown remarkable success in uncertainty estimation. However, there is still room for improvement, particularly in out-of-distribution (OOD) detection and classification tasks. The limited OOD detection performance of EDL arises from its inability to reflect the distance between the testing example and training data when quantifying uncertainty, while its limited classification performance stems from its parameterization of the concentration parameters. To address these limitations, we propose a novel method called Density Aware Evidential Deep Learning (DAEDL). DAEDL integrates the feature space density of the testing example with the output of EDL during the prediction stage, while using a novel parameterization that resolves the issues in the conventional parameterization. We prove that DAEDL enjoys a number of favorable theoretical properties. DAEDL demonstrates state-of-the-art performance across diverse downstream tasks related to uncertainty estimation and classification.
|
https://proceedings.mlr.press/v235/yoon24b.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yoon24b/yoon24b.pdf
|
https://openreview.net/forum?id=ZeF75iQcAc
|
Optimal Acceleration for Minimax and Fixed-Point Problems is Not Unique
|
https://proceedings.mlr.press/v235/yoon24b.html
|
Taeho Yoon, Jaeyeon Kim, Jaewook J. Suh, Ernest K. Ryu
|
https://proceedings.mlr.press/v235/yoon24b.html
|
ICML 2024
|
Recently, accelerated algorithms using the anchoring mechanism for minimax optimization and fixed-point problems have been proposed, and matching complexity lower bounds establish their optimality. In this work, we present the surprising observation that the optimal acceleration mechanism in minimax optimization and fixed-point problems is not unique. Our new algorithms achieve exactly the same worst-case convergence rates as existing anchor-based methods while using materially different acceleration mechanisms. Specifically, these new algorithms are dual to the prior anchor-based accelerated methods in the sense of H-duality. This finding opens a new avenue of research on accelerated algorithms since we now have a family of methods that empirically exhibit varied characteristics while having the same optimal worst-case guarantee.
|
https://proceedings.mlr.press/v235/yoon24c.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yoon24c/yoon24c.pdf
|
https://openreview.net/forum?id=OnEaBGU3LO
|
FRAG: Frequency Adapting Group for Diffusion Video Editing
|
https://proceedings.mlr.press/v235/yoon24c.html
|
Sunjae Yoon, Gwanhyeong Koo, Geonwoo Kim, Chang D. Yoo
|
https://proceedings.mlr.press/v235/yoon24c.html
|
ICML 2024
|
In video editing, the hallmark of a quality edit lies in its consistent and unobtrusive adjustment. Modification, when integrated, must be smooth and subtle, preserving the natural flow and aligning seamlessly with the original vision. Therefore, our primary focus is on overcoming the current challenges in high quality edit to ensure that each edit enhances the final product without disrupting its intended essence. However, quality deterioration such as blurring and flickering is routinely observed in recent diffusion video editing systems. We confirm that this deterioration often stems from high-frequency leak: the diffusion model fails to accurately synthesize high-frequency components during denoising process. To this end, we devise Frequency Adapting Group (FRAG) which enhances the video quality in terms of consistency and fidelity by introducing a novel receptive field branch to preserve high-frequency components during the denoising process. FRAG is performed in a model-agnostic manner without additional training and validates the effectiveness on video editing benchmarks (i.e., TGVE, DAVIS).
|
https://proceedings.mlr.press/v235/yoon24d.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yoon24d/yoon24d.pdf
|
https://openreview.net/forum?id=59MYoLghyk
|
Breadth-First Exploration on Adaptive Grid for Reinforcement Learning
|
https://proceedings.mlr.press/v235/yoon24d.html
|
Youngsik Yoon, Gangbok Lee, Sungsoo Ahn, Jungseul Ok
|
https://proceedings.mlr.press/v235/yoon24d.html
|
ICML 2024
|
Graph-based planners have gained significant attention for goal-conditioned reinforcement learning (RL), where they construct a graph consisting of confident transitions between subgoals as edges and run shortest path algorithms to exploit the confident edges. Meanwhile, identifying and avoiding unattainable transitions are also crucial yet overlooked by the previous graph-based planners, leading to wasting an excessive number of attempts at unattainable subgoals. To address this oversight, we propose a graph construction method that efficiently manages all the achieved and unattained subgoals on a grid graph adaptively discretizing the goal space. This enables a breadth-first exploration strategy, grounded in the local adaptive grid refinement, that prioritizes broad probing of subgoals on a coarse grid over meticulous one on a dense grid. We conducted a theoretical analysis and demonstrated the effectiveness of our approach through empirical evidence, showing that only BEAG succeeds in complex environments under the proposed fixed-goal setting.
|
https://proceedings.mlr.press/v235/you24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/you24a/you24a.pdf
|
https://openreview.net/forum?id=7mFSaP6IiN
|
When Linear Attention Meets Autoregressive Decoding: Towards More Effective and Efficient Linearized Large Language Models
|
https://proceedings.mlr.press/v235/you24a.html
|
Haoran You, Yichao Fu, Zheng Wang, Amir Yazdanbakhsh, Yingyan Celine Lin
|
https://proceedings.mlr.press/v235/you24a.html
|
ICML 2024
|
Autoregressive Large Language Models (LLMs) have achieved impressive performance in language tasks but face two significant bottlenecks: (1) quadratic complexity in the attention module as the number of tokens increases, and (2) limited efficiency due to the sequential processing nature of autoregressive LLMs during generation. While linear attention and speculative decoding offer potential solutions, their applicability and synergistic potential for enhancing autoregressive LLMs remain uncertain. We conduct the first comprehensive study on the efficacy of existing linear attention methods for autoregressive LLMs, integrating them with speculative decoding. We introduce an augmentation technique for linear attention that ensures compatibility with speculative decoding, enabling more efficient training and serving of LLMs. Extensive experiments and ablation studies involving seven existing linear attention models and five encoder/decoder-based LLMs consistently validate the effectiveness of our augmented linearized LLMs. Notably, our approach achieves up to a 6.67 reduction in perplexity on the LLaMA model and up to a 2$\times$ speedup during generation compared to prior linear attention methods. Codes and models are available at https://github.com/GATECH-EIC/Linearized-LLM.
|
https://proceedings.mlr.press/v235/you24b.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/you24b/you24b.pdf
|
https://openreview.net/forum?id=NeotatlYOL
|
SpikeZIP-TF: Conversion is All You Need for Transformer-based SNN
|
https://proceedings.mlr.press/v235/you24b.html
|
Kang You, Zekai Xu, Chen Nie, Zhijie Deng, Qinghai Guo, Xiang Wang, Zhezhi He
|
https://proceedings.mlr.press/v235/you24b.html
|
ICML 2024
|
Spiking neural network (SNN) has attracted great attention due to its characteristic of high efficiency and accuracy. Currently, the ANN-to-SNN conversion methods can obtain ANN on-par accuracy SNN with ultra-low latency (8 time-steps) in CNN structure on computer vision (CV) tasks. However, as Transformer-based networks have achieved prevailing precision on both CV and natural language processing (NLP), the Transformer-based SNNs are still encounting the lower accuracy w.r.t the ANN counterparts. In this work, we introduce a novel ANN-to-SNN conversion method called SpikeZIP-TF, where ANN and SNN are exactly equivalent, thus incurring no accuracy degradation. SpikeZIP-TF achieves 83.82% accuracy on CV dataset (ImageNet) and 93.79% accuracy on NLP dataset (SST-2), which are higher than SOTA Transformer-based SNNs. The code is available in GitHub: https://github.com/Intelligent-Computing-Research-Group/SpikeZIP_transformer
|
https://proceedings.mlr.press/v235/yu24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24a/yu24a.pdf
|
https://openreview.net/forum?id=pOJbk4Nzmi
|
Efficient Algorithms for Empirical Group Distributionally Robust Optimization and Beyond
|
https://proceedings.mlr.press/v235/yu24a.html
|
Dingzhi Yu, Yunuo Cai, Wei Jiang, Lijun Zhang
|
https://proceedings.mlr.press/v235/yu24a.html
|
ICML 2024
|
In this paper, we investigate the empirical counterpart of Group Distributionally Robust Optimization (GDRO), which aims to minimize the maximal empirical risk across $m$ distinct groups. We formulate empirical GDRO as a two-level finite-sum convex-concave minimax optimization problem and develop an algorithm called ALEG to benefit from its special structure. ALEG is a double-looped stochastic primal-dual algorithm that incorporates variance reduction techniques into a modified mirror prox routine. To exploit the two-level finite-sum structure, we propose a simple group sampling strategy to construct the stochastic gradient with a smaller Lipschitz constant and then perform variance reduction for all groups. Theoretical analysis shows that ALEG achieves $\varepsilon$-accuracy within a computation complexity of $\mathcal{O}\left(\frac{m\sqrt{\bar{n}\ln{m}}}{\varepsilon}\right)$, where $\bar n$ is the average number of samples among $m$ groups. Notably, our approach outperforms the state-of-the-art method by a factor of $\sqrt{m}$. Based on ALEG, we further develop a two-stage optimization algorithm called ALEM to deal with the empirical Minimax Excess Risk Optimization (MERO) problem. The computation complexity of ALEM nearly matches that of ALEG, surpassing the rates of existing methods.
|
https://proceedings.mlr.press/v235/yu24b.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24b/yu24b.pdf
|
https://openreview.net/forum?id=O45u81aby2
|
Towards Resource-friendly, Extensible and Stable Incomplete Multi-view Clustering
|
https://proceedings.mlr.press/v235/yu24b.html
|
Shengju Yu, Zhibin Dong, Siwei Wang, Xinhang Wan, Yue Liu, Weixuan Liang, Pei Zhang, Wenxuan Tu, Xinwang Liu
|
https://proceedings.mlr.press/v235/yu24b.html
|
ICML 2024
|
Incomplete multi-view clustering (IMVC) methods typically encounter three drawbacks: (1) intense time and/or space overheads; (2) intractable hyper-parameters; (3) non-zero variance results. With these concerns in mind, we give a simple yet effective IMVC scheme, termed as ToRES. Concretely, instead of self-expression affinity, we manage to construct prototype-sample affinity for incomplete data so as to decrease the memory requirements. To eliminate hyper-parameters, besides mining complementary features among views by view-wise prototypes, we also attempt to devise cross-view prototypes to capture consensus features for jointly forming high-quality clustering representation. To avoid the variance, we successfully unify representation learning and clustering operation, and directly optimize the discrete cluster indicators from incomplete data. Then, for the resulting objective function, we provide two equivalent solutions from perspectives of feasible region partitioning and objective transformation. Many results suggest that ToRES exhibits advantages against 20 SOTA algorithms, even in scenarios with a higher ratio of incomplete data.
|
https://proceedings.mlr.press/v235/yu24c.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24c/yu24c.pdf
|
https://openreview.net/forum?id=IArWwIim8M
|
Activation-Descent Regularization for Input Optimization of ReLU Networks
|
https://proceedings.mlr.press/v235/yu24c.html
|
Hongzhan Yu, Sicun Gao
|
https://proceedings.mlr.press/v235/yu24c.html
|
ICML 2024
|
We present a new approach for input optimization of ReLU networks that explicitly takes into account the effect of changes in activation patterns. We analyze local optimization steps in both the input space and the space of activation patterns to propose methods with superior local descent properties. To accomplish this, we convert the discrete space of activation patterns into differentiable representations and propose regularization terms that improve each descent step. Our experiments demonstrate the effectiveness of the proposed input-optimization methods for improving the state-of-the-art in various areas, such as adversarial learning, generative modeling, and reinforcement learning.
|
https://proceedings.mlr.press/v235/yu24d.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24d/yu24d.pdf
|
https://openreview.net/forum?id=LkJ6qOMv77
|
Collage: Light-Weight Low-Precision Strategy for LLM Training
|
https://proceedings.mlr.press/v235/yu24d.html
|
Tao Yu, Gaurav Gupta, Karthick Gopalswamy, Amith R Mamidala, Hao Zhou, Jeffrey Huynh, Youngsuk Park, Ron Diamant, Anoop Deoras, Luke Huan
|
https://proceedings.mlr.press/v235/yu24d.html
|
ICML 2024
|
Large models training is plagued by the intense compute cost and limited hardware memory. A practical solution is low-precision representation but is troubled by loss in numerical accuracy and unstable training rendering the model less useful. We argue that low-precision floating points can perform well provided the error is properly compensated at the critical locations in the training process. We propose Collage which utilizes multi-component float representation in low-precision to accurately perform operations with numerical errors accounted. To understand the impact of imprecision to training, we propose a simple and novel metric which tracks the lost information during training as well as differentiates various precision strategies. Our method works with commonly used low-precision such as half-precision ($16$-bit floating points) and can be naturally extended to work with even lower precision such as $8$-bit. Experimental results show that pre-training using Collage removes the requirement of using $32$-bit floating-point copies of the model and attains similar/better training performance compared to $(16, 32)$-bit mixed-precision strategy, with up to $3.7\times$ speedup and $\sim 15%$ to $23%$ less memory usage in practice. The code is available at https://github.com/amazon-science/collage.
|
https://proceedings.mlr.press/v235/yu24e.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24e/yu24e.pdf
|
https://openreview.net/forum?id=mUT1biz09t
|
Privacy-Preserving Instructions for Aligning Large Language Models
|
https://proceedings.mlr.press/v235/yu24e.html
|
Da Yu, Peter Kairouz, Sewoong Oh, Zheng Xu
|
https://proceedings.mlr.press/v235/yu24e.html
|
ICML 2024
|
Service providers of large language model (LLM) applications collect user instructions in the wild and use them in further aligning LLMs with users’ intentions. These instructions, which potentially contain sensitive information, are annotated by human workers in the process. This poses a new privacy risk not addressed by the typical private optimization. To this end, we propose using synthetic instructions to replace real instructions in data annotation and model fine-tuning. Formal differential privacy is guaranteed by generating those synthetic instructions using privately fine-tuned generators. Crucial in achieving the desired utility is our novel filtering algorithm that matches the distribution of the synthetic instructions to that of the real ones. In both supervised fine-tuning and reinforcement learning from human feedback, our extensive experiments demonstrate the high utility of the final set of synthetic instructions by showing comparable results to real instructions. In supervised fine-tuning, models trained with private synthetic instructions outperform leading open-source models such as Vicuna.
|
https://proceedings.mlr.press/v235/yu24f.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24f/yu24f.pdf
|
https://openreview.net/forum?id=kKWjZoaRLv
|
Learning Latent Structures in Network Games via Data-Dependent Gated-Prior Graph Variational Autoencoders
|
https://proceedings.mlr.press/v235/yu24f.html
|
Xue Yu, Muchen Li, Yan Leng, Renjie Liao
|
https://proceedings.mlr.press/v235/yu24f.html
|
ICML 2024
|
In network games, individuals interact strategically within network environments to maximize their utilities. However, obtaining network structures is challenging. In this work, we propose an unsupervised learning model, called data-dependent gated-prior graph variational autoencoder (GPGVAE), that infers the underlying latent interaction type (strategic complement vs. substitute) among individuals and the latent network structure based on their observed actions. Specially, we propose a spectral graph neural network (GNN) based encoder to predict the interaction type and a data-dependent gated prior that models network structures conditioned on the interaction type. We further propose a Transformer based mixture of Bernoulli encoder of network structures and a GNN based decoder of game actions. We systematically study the Monte Carlo gradient estimation methods and effectively train our model in a stage-wise fashion. Extensive experiments across various synthetic and real-world network games demonstrate that our model achieves state-of-the-art performances in inferring network structures and well captures interaction types.
|
https://proceedings.mlr.press/v235/yu24g.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24g/yu24g.pdf
|
https://openreview.net/forum?id=udFZhUgtkI
|
Learning Scale-Aware Spatio-temporal Implicit Representation for Event-based Motion Deblurring
|
https://proceedings.mlr.press/v235/yu24g.html
|
Wei Yu, Jianing Li, Shengping Zhang, Xiangyang Ji
|
https://proceedings.mlr.press/v235/yu24g.html
|
ICML 2024
|
Existing event-based motion deblurring methods mostly focus on restoring images with the same spatial and temporal scales as events. However, the unknown scales of images and events in the real world pose great challenges and have rarely been explored. To address this gap, we propose a novel Scale-Aware Spatio-temporal Network (SASNet) to flexibly restore blurred images with event streams at arbitrary scales. The core idea is to implicitly aggregate both spatial and temporal correspondence features of images and events to generalize at continuous scales. To restore highly blurred local areas, we develop a Spatial Implicit Representation Module (SIRM) to aggregate spatial correlation at any resolution through event encoding sampling. To tackle global motion blur, a Temporal Implicit Representation Module (TIRM) is presented to learn temporal correlation via temporal shift operations with long-term aggregation. Additionally, we build a High-resolution Hybrid Deblur (H2D) dataset using a new-generation hybrid event-based sensor, which comprises images with naturally spatially aligned and temporally synchronized events at various scales. Experiments demonstrate that our SASNet outperforms state-of-the-art methods on both synthetic GoPro and real H2D datasets, especially in high-speed motion scenarios. Code and dataset are available at https://github.com/aipixel/SASNet.
|
https://proceedings.mlr.press/v235/yu24h.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24h/yu24h.pdf
|
https://openreview.net/forum?id=LabSWooau0
|
Enabling Few-Shot Learning with PID Control: A Layer Adaptive Optimizer
|
https://proceedings.mlr.press/v235/yu24h.html
|
Le Yu, Xinde Li, Pengfei Zhang, Zhentong Zhang, Fir Dunkin
|
https://proceedings.mlr.press/v235/yu24h.html
|
ICML 2024
|
Model-Agnostic Meta-Learning (MAML) and its variants have shown remarkable performance in scenarios characterized by a scarcity of labeled data during the training phase of machine learning models. Despite these successes, MAMLbased approaches encounter significant challenges when there is a substantial discrepancy in the distribution of training and testing tasks, resulting in inefficient learning and limited generalization across domains. Inspired by classical proportional-integral-derivative (PID) control theory, this study introduces a Layer-Adaptive PID (LA-PID) Optimizer, a MAML-based optimizer that employs efficient parameter optimization methods to dynamically adjust task-specific PID control gains at each layer of the network, conducting a first-principles analysis of optimal convergence conditions. A series of experiments conducted on four standard benchmark datasets demonstrate the efficacy of the LA-PID optimizer, indicating that LA-PID achieves state-oftheart performance in few-shot classification and cross-domain tasks, accomplishing these objectives with fewer training steps. Code is available on https://github.com/yuguopin/LA-PID.
|
https://proceedings.mlr.press/v235/yu24i.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24i/yu24i.pdf
|
https://openreview.net/forum?id=ZdqiT0McON
|
Generalization Bound and New Algorithm for Clean-Label Backdoor Attack
|
https://proceedings.mlr.press/v235/yu24i.html
|
Lijia Yu, Shuang Liu, Yibo Miao, Xiao-Shan Gao, Lijun Zhang
|
https://proceedings.mlr.press/v235/yu24i.html
|
ICML 2024
|
The generalization bound is a crucial theoretical tool for assessing the generalizability of learning methods and there exist vast literatures on generalizability of normal learning, adversarial learning, and data poisoning. Unlike other data poison attacks, the backdoor attack has the special property that the poisoned triggers are contained in both the training set and the test set and the purpose of the attack is two-fold. To our knowledge, the generalization bound for the backdoor attack has not been established. In this paper, we fill this gap by deriving algorithm-independent generalization bounds in the clean-label backdoor attack scenario. Precisely, based on the goals of backdoor attack, we give upper bounds for the clean sample population errors and the poison population errors in terms of the empirical error on the poisoned training dataset. Furthermore, based on the theoretical result, a new clean-label backdoor attack is proposed that computes the poisoning trigger by combining adversarial noise and indiscriminate poison. We show its effectiveness in a variety of settings.
|
https://proceedings.mlr.press/v235/yu24j.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24j/yu24j.pdf
|
https://openreview.net/forum?id=HkWxjpUV0S
|
Learning Causal Dynamics Models in Object-Oriented Environments
|
https://proceedings.mlr.press/v235/yu24j.html
|
Zhongwei Yu, Jingqing Ruan, Dengpeng Xing
|
https://proceedings.mlr.press/v235/yu24j.html
|
ICML 2024
|
Causal dynamics models (CDMs) have demonstrated significant potential in addressing various challenges in reinforcement learning. To learn CDMs, recent studies have performed causal discovery to capture the causal dependencies among environmental variables. However, the learning of CDMs is still confined to small-scale environments due to computational complexity and sample efficiency constraints. This paper aims to extend CDMs to large-scale object-oriented environments, which consist of a multitude of objects classified into different categories. We introduce the Object-Oriented CDM (OOCDM) that shares causalities and parameters among objects belonging to the same class. Furthermore, we propose a learning method for OOCDM that enables it to adapt to a varying number of objects. Experiments on large-scale tasks indicate that OOCDM outperforms existing CDMs in terms of causal discovery, prediction accuracy, generalization, and computational efficiency.
|
https://proceedings.mlr.press/v235/yu24k.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24k/yu24k.pdf
|
https://openreview.net/forum?id=6aKwVmHQI1
|
ViP: A Differentially Private Foundation Model for Computer Vision
|
https://proceedings.mlr.press/v235/yu24k.html
|
Yaodong Yu, Maziar Sanjabi, Yi Ma, Kamalika Chaudhuri, Chuan Guo
|
https://proceedings.mlr.press/v235/yu24k.html
|
ICML 2024
|
Artificial intelligence (AI) has seen a tremendous surge in capabilities thanks to the use of foundation models trained on internet-scale data. On the flip side, the uncurated nature of internet-scale data also poses significant privacy and legal risks, as they often contain personal information or copyrighted material that should not be trained on without permission. In this work, we propose as a mitigation measure a recipe to train foundation vision models via self-supervised learning with differential privacy (DP) guarantee. We identify masked autoencoders as a suitable learning algorithm that aligns well with DP-SGD, and train ViP—a Vision transformer with differential Privacy—under a strict privacy budget of $\epsilon=8$ on the LAION400M dataset. We evaluate the quality of representation learned by ViP using standard downstream vision tasks; in particular, ViP achieves a (non-private) linear probing accuracy of 55.7% on ImageNet, comparable to that of end-to-end trained AlexNet (trained and evaluated on ImageNet). Our result suggests that scaling to internet-scale data can be practical for private learning. Code and DP pre-trained models are available at https://github.com/facebookresearch/ViP-MAE.
|
https://proceedings.mlr.press/v235/yu24l.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24l/yu24l.pdf
|
https://openreview.net/forum?id=DLTjFFiuUJ
|
Unveiling and Harnessing Hidden Attention Sinks: Enhancing Large Language Models without Training through Attention Calibration
|
https://proceedings.mlr.press/v235/yu24l.html
|
Zhongzhi Yu, Zheng Wang, Yonggan Fu, Huihong Shi, Khalid Shaikh, Yingyan Celine Lin
|
https://proceedings.mlr.press/v235/yu24l.html
|
ICML 2024
|
Attention is a fundamental component behind the remarkable achievements of large language models (LLMs). However, our current understanding of the attention mechanism, especially regarding how attention distributions are established, remains limited. Inspired by recent studies that explore the presence of attention sink in the initial token, which receives disproportionately large attention scores despite their lack of semantic importance, this work delves deeper into this phenomenon. We aim to provide a more profound understanding of the existence of attention sinks within LLMs and to uncover ways to enhance the achievable accuracy of LLMs by directly optimizing the attention distributions, without the need for weight finetuning. Specifically, this work begins with comprehensive visualizations of the attention distributions in LLMs during inference across various inputs and tasks. Based on these visualizations, to the best of our knowledge, we are the first to discover that (1) attention sinks occur not only at the start of sequences but also within later tokens of the input, and (2) not all attention sinks have a positive impact on the achievable accuracy of LLMs. Building upon our findings, we propose a training-free Attention Calibration Technique (ACT) that automatically optimizes the attention distributions on the fly during inference in an input-adaptive manner. Extensive experiments validate that ACT consistently enhances the accuracy of various LLMs across different applications. Specifically, ACT achieves an average improvement of up to $7.30%$ in accuracy across different datasets when applied to Llama-30B.
|
https://proceedings.mlr.press/v235/yu24m.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24m/yu24m.pdf
|
https://openreview.net/forum?id=0LBNdbmQCM
|
Purify Unlearnable Examples via Rate-Constrained Variational Autoencoders
|
https://proceedings.mlr.press/v235/yu24m.html
|
Yi Yu, Yufei Wang, Song Xia, Wenhan Yang, Shijian Lu, Yap-Peng Tan, Alex Kot
|
https://proceedings.mlr.press/v235/yu24m.html
|
ICML 2024
|
Unlearnable examples (UEs) seek to maximize testing error by making subtle modifications to training examples that are correctly labeled. Defenses against these poisoning attacks can be categorized based on whether specific interventions are adopted during training. The first approach is training-time defense, such as adversarial training, which can mitigate poisoning effects but is computationally intensive. The other approach is pre-training purification, e.g., image short squeezing, which consists of several simple compressions but often encounters challenges in dealing with various UEs. Our work provides a novel disentanglement mechanism to build an efficient pre-training purification method. Firstly, we uncover rate-constrained variational autoencoders (VAEs), demonstrating a clear tendency to suppress the perturbations in UEs. We subsequently conduct a theoretical analysis for this phenomenon. Building upon these insights, we introduce a disentangle variational autoencoder (D-VAE), capable of disentangling the perturbations with learnable class-wise embeddings. Based on this network, a two-stage purification approach is naturally developed. The first stage focuses on roughly eliminating perturbations, while the second stage produces refined, poison-free results, ensuring effectiveness and robustness across various scenarios. Extensive experiments demonstrate the remarkable performance of our method across CIFAR-10, CIFAR-100, and a 100-class ImageNet-subset. Code is available at https://github.com/yuyi-sd/D-VAE.
|
https://proceedings.mlr.press/v235/yu24n.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24n/yu24n.pdf
|
https://openreview.net/forum?id=ZZ7UKgK4c1
|
Few-Shot Character Understanding in Movies as an Assessment to Meta-Learning of Theory-of-Mind
|
https://proceedings.mlr.press/v235/yu24n.html
|
Mo Yu, Qiujing Wang, Shunchi Zhang, Yisi Sang, Kangsheng Pu, Zekai Wei, Han Wang, Liyan Xu, Jing Li, Yue Yu, Jie Zhou
|
https://proceedings.mlr.press/v235/yu24n.html
|
ICML 2024
|
When reading a story, humans can quickly understand new fictional characters with a few observations, mainly by drawing analogies to fictional and real people they already know. This reflects the few-shot and meta-learning essence of humans’ inference of characters’ mental states, i.e., theory-of-mind (ToM), which is largely ignored in existing research. We fill this gap with a novel NLP dataset in a realistic narrative understanding scenario, ToM-in-AMC. Our dataset consists of $\sim$1,000 parsed movie scripts, each corresponding to a few-shot character understanding task that requires models to mimic humans’ ability of fast digesting characters with a few starting scenes in a new movie. We further propose a novel ToM prompting approach designed to explicitly assess the influence of multiple ToM dimensions. It surpasses existing baseline models, underscoring the significance of modeling multiple ToM dimensions for our task. Our extensive human study verifies that humans are capable of solving our problem by inferring characters’ mental states based on their previously seen movies. In comparison, all the AI systems lag $>20%$ behind humans, highlighting a notable limitation in existing approaches’ ToM capabilities. Code and data are available at https://github.com/ShunchiZhang/ToM-in-AMC
|
https://proceedings.mlr.press/v235/yu24o.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24o/yu24o.pdf
|
https://openreview.net/forum?id=KOTutrSR2y
|
MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities
|
https://proceedings.mlr.press/v235/yu24o.html
|
Weihao Yu, Zhengyuan Yang, Linjie Li, Jianfeng Wang, Kevin Lin, Zicheng Liu, Xinchao Wang, Lijuan Wang
|
https://proceedings.mlr.press/v235/yu24o.html
|
ICML 2024
|
We propose MM-Vet, an evaluation benchmark that examines large multimodal models (LMMs) on complicated multimodal tasks. Recent LMMs have shown various intriguing abilities, such as solving math problems written on the blackboard, reasoning about events and celebrities in news images, and explaining visual jokes. Rapid model advancements pose challenges to evaluation benchmark development. Problems include: (1) How to systematically structure and evaluate the complicated multimodal tasks; (2) How to design evaluation metrics that work well across question and answer types; and (3) How to give model insights beyond a simple performance ranking. To this end, we present MM-Vet, designed based on the insight that the intriguing ability to solve complicated tasks is often achieved by a generalist model being able to integrate different core vision-language (VL) capabilities. MM-Vet defines 6 core VL capabilities and examines the 16 integrations of interest derived from the capability combination. For evaluation metrics, we propose an LLM-based evaluator for open-ended outputs. The evaluator enables the evaluation across different question types and answer styles, resulting in a unified scoring metric. We evaluate representative LMMs on MM-Vet, providing insights into the capabilities of different LMM system paradigms and models.
|
https://proceedings.mlr.press/v235/yu24p.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24p/yu24p.pdf
|
https://openreview.net/forum?id=fq0NaiU8Ex
|
Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch
|
https://proceedings.mlr.press/v235/yu24p.html
|
Le Yu, Bowen Yu, Haiyang Yu, Fei Huang, Yongbin Li
|
https://proceedings.mlr.press/v235/yu24p.html
|
ICML 2024
|
In this paper, we unveil that Language Models (LMs) can acquire new capabilities by assimilating parameters from homologous models without retraining or GPUs. We first introduce DARE to set most delta parameters (i.e., the disparity between fine-tuned and pre-trained parameters) to zeros without affecting the abilities of Supervised Fine-Tuning (SFT) LMs, which randomly Drops delta parameters with a ratio $p$ And REscales the remaining ones by $1 / (1 - p)$ to approximate the original embeddings. Then, we use DARE as a versatile plug-in to sparsify delta parameters of multiple SFT homologous models for mitigating parameter interference and merge them into a single model by parameter fusing. We experiment with encoder- and decoder-based LMs, showing that: (1) SFT delta parameter value ranges are typically small (within 0.002) with extreme redundancy, and DARE can effortlessly eliminate 90% or even 99% of them; (2) DARE can merge multiple task-specific LMs into one LM with diverse capabilities. Notably, this phenomenon is more pronounced in large-scale LMs, where the merged LM reveals the potential to surpass the performance of any source LM, providing a new discovery. We also utilize DARE to create a merged LM that ranks first among models with 7 billion parameters on the Open LLM Leaderboard.
|
https://proceedings.mlr.press/v235/yu24q.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24q/yu24q.pdf
|
https://openreview.net/forum?id=J9YKDvqr65
|
Improving Sharpness-Aware Minimization by Lookahead
|
https://proceedings.mlr.press/v235/yu24q.html
|
Runsheng Yu, Youzhi Zhang, James Kwok
|
https://proceedings.mlr.press/v235/yu24q.html
|
ICML 2024
|
Sharpness-Aware Minimization (SAM), which performs gradient descent on adversarially perturbed weights, can improve generalization by identifying flatter minima. However, recent studies have shown that SAM may suffer from convergence instability and oscillate around saddle points, resulting in slow convergence and inferior performance. To address this problem, we propose the use of a lookahead mechanism to gather more information about the landscape by looking further ahead, and thus find a better trajectory to converge. By examining the nature of SAM, we simplify the extrapolation procedure, resulting in a more efficient algorithm. Theoretical results show that the proposed method converges to a stationary point and is less prone to saddle points. Experiments on standard benchmark datasets also verify that the proposed method outperforms the SOTAs, and converge more effectively to flat minima.
|
https://proceedings.mlr.press/v235/yu24r.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24r/yu24r.pdf
|
https://openreview.net/forum?id=ZFRrOiZruJ
|
A Unified Adaptive Testing System Enabled by Hierarchical Structure Search
|
https://proceedings.mlr.press/v235/yu24r.html
|
Junhao Yu, Yan Zhuang, Zhenya Huang, Qi Liu, Xin Li, Rui Li, Enhong Chen
|
https://proceedings.mlr.press/v235/yu24r.html
|
ICML 2024
|
Adaptive Testing System (ATS) is a promising testing mode, extensively utilized in standardized tests like the GRE. It offers personalized ability assessment by dynamically adjusting questions based on individual ability levels. Compared to traditional exams, ATS can improve the accuracy of ability estimates while simultaneously reducing the number of questions required. Despite the diverse testing formats of ATS, tailored to different adaptability requirements in various testing scenarios, there is a notable absence of a unified framework for modeling them. In this paper, we introduce a unified data-driven ATS framework that conceptualizes the various testing formats as a hierarchical test structure search problem. It can learn directly from data to solve for the optimal questions for each student, eliminating the need for manual test design. The proposed solution algorithm comes with theoretical guarantees for estimation error and convergence. Empirical results show that our framework maintains assessment accuracy while reducing question count by 20% on average and improving training stability.
|
https://proceedings.mlr.press/v235/yu24s.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yu24s/yu24s.pdf
|
https://openreview.net/forum?id=87CYNyCGOo
|
Revitalizing Multivariate Time Series Forecasting: Learnable Decomposition with Inter-Series Dependencies and Intra-Series Variations Modeling
|
https://proceedings.mlr.press/v235/yu24s.html
|
Guoqi Yu, Jing Zou, Xiaowei Hu, Angelica I Aviles-Rivero, Jing Qin, Shujun Wang
|
https://proceedings.mlr.press/v235/yu24s.html
|
ICML 2024
|
Predicting multivariate time series is crucial, demanding precise modeling of intricate patterns, including inter-series dependencies and intra-series variations. Distinctive trend characteristics in each time series pose challenges, and existing methods, relying on basic moving average kernels, may struggle with the non-linear structure and complex trends in real-world data. Given that, we introduce a learnable decomposition strategy to capture dynamic trend information more reasonably. Additionally, we propose a dual attention module tailored to capture inter-series dependencies and intra-series variations simultaneously for better time series forecasting, which is implemented by channel-wise self-attention and autoregressive self-attention. To evaluate the effectiveness of our method, we conducted experiments across eight open-source datasets and compared it with the state-of-the-art methods. Through the comparison results, our $\textbf{Leddam}$ ($\textbf{LE}arnable$ $\textbf{D}ecomposition$ and $\textbf{D}ual $ $\textbf{A}ttention$ $\textbf{M}odule$) not only demonstrates significant advancements in predictive performance but also the proposed decomposition strategy can be plugged into other methods with a large performance-boosting, from 11.87% to 48.56% MSE error degradation. Code is available at this link: https://github.com/Levi-Ackman/Leddam.
|
https://proceedings.mlr.press/v235/yuan24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yuan24a/yuan24a.pdf
|
https://openreview.net/forum?id=Zs3qW8Njov
|
Smoothing Proximal Gradient Methods for Nonsmooth Sparsity Constrained Optimization: Optimality Conditions and Global Convergence
|
https://proceedings.mlr.press/v235/yuan24a.html
|
Ganzhao Yuan
|
https://proceedings.mlr.press/v235/yuan24a.html
|
ICML 2024
|
Nonsmooth sparsity constrained optimization encompasses a broad spectrum of applications in machine learning. This problem is generally non-convex and NP-hard. Existing solutions to this problem exhibit several notable limitations, including their inability to address general nonsmooth problems, tendency to yield weaker optimality conditions, and lack of comprehensive convergence analysis. This paper considers Smoothing Proximal Gradient Methods (SPGM) as solutions to nonsmooth sparsity constrained optimization problems. Two specific variants of SPGM are explored: one based on Iterative Hard Thresholding (SPGM-IHT) and the other on Block Coordinate Decomposition (SPGM-BCD). It is shown that the SPGM-BCD algorithm finds stronger stationary points compared to previous methods. Additionally, novel theories for analyzing the convergence rates to approximate global optimal solutions of both the SPGM-IHT and SPGM-BCD algorithms are developed. Our theoretical bounds, capitalizing on the intrinsic sparsity of the optimization problem, are on par with the best-known error bounds available to date. Finally, numerical experiments reveal that SPGM-IHT performs comparably to current IHT-style methods, while SPGM-BCD consistently surpasses them.
|
https://proceedings.mlr.press/v235/yuan24b.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yuan24b/yuan24b.pdf
|
https://openreview.net/forum?id=oYltxxam2t
|
A Linear Time and Space Local Point Cloud Geometry Encoder via Vectorized Kernel Mixture (VecKM)
|
https://proceedings.mlr.press/v235/yuan24b.html
|
Dehao Yuan, Cornelia Fermuller, Tahseen Rabbani, Furong Huang, Yiannis Aloimonos
|
https://proceedings.mlr.press/v235/yuan24b.html
|
ICML 2024
|
We propose VecKM, a local point cloud geometry encoder that is descriptive and efficient to compute. VecKM leverages a unique approach by vectorizing a kernel mixture to represent the local point cloud. Such representation’s descriptiveness is supported by two theorems that validate its ability to reconstruct and preserve the similarity of the local shape. Unlike existing encoders downsampling the local point cloud, VecKM constructs the local geometry encoding using all neighboring points, producing a more descriptive encoding. Moreover, VecKM is efficient to compute and scalable to large point cloud inputs: VecKM reduces the memory cost from $(n^2+nKd)$ to $(nd+np)$; and reduces the major runtime cost from computing $nK$ MLPs to $n$ MLPs, where $n$ is the size of the point cloud, $K$ is the neighborhood size, $d$ is the encoding dimension, and $p$ is a marginal factor. The efficiency is due to VecKM’s unique factorizable property that eliminates the need of explicitly grouping points into neighbors. In the normal estimation task, VecKM demonstrates not only 100x faster inference speed but also highest accuracy and strongest robustness. In classification and segmentation tasks, integrating VecKM as a preprocessing module achieves consistently better performance than the PointNet, PointNet++, and point transformer baselines, and runs consistently faster by up to 10 times.
|
https://proceedings.mlr.press/v235/yuan24c.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yuan24c/yuan24c.pdf
|
https://openreview.net/forum?id=nMWxLnSBGW
|
SHINE: Shielding Backdoors in Deep Reinforcement Learning
|
https://proceedings.mlr.press/v235/yuan24c.html
|
Zhuowen Yuan, Wenbo Guo, Jinyuan Jia, Bo Li, Dawn Song
|
https://proceedings.mlr.press/v235/yuan24c.html
|
ICML 2024
|
Recent studies have discovered that a deep reinforcement learning (DRL) policy is vulnerable to backdoor attacks. Existing defenses against backdoor attacks either do not consider RL’s unique mechanism or make unrealistic assumptions, resulting in limited defense efficacy, practicability, and generalizability. We propose SHINE, a backdoor shielding method specific for DRL. SHINE designs novel policy explanation techniques to identify the backdoor triggers and a policy retraining algorithm to eliminate the impact of the triggers on backdoored agents. We theoretically justify that SHINE guarantees to improve a backdoored agent’s performance in a poisoned environment while ensuring its performance difference in the clean environment before and after shielding is bounded. We further conduct extensive experiments that evaluate SHINE against three mainstream DRL backdoor attacks in various benchmark RL environments. Our results show that SHINE significantly outperforms existing defenses in mitigating these backdoor attacks.
|
https://proceedings.mlr.press/v235/yuan24d.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yuan24d/yuan24d.pdf
|
https://openreview.net/forum?id=0NphYCmgua
|
Self-Rewarding Language Models
|
https://proceedings.mlr.press/v235/yuan24d.html
|
Weizhe Yuan, Richard Yuanzhe Pang, Kyunghyun Cho, Xian Li, Sainbayar Sukhbaatar, Jing Xu, Jason E Weston
|
https://proceedings.mlr.press/v235/yuan24d.html
|
ICML 2024
|
We posit that to achieve superhuman agents, future models require superhuman feedback in order to provide an adequate training signal. Current approaches commonly train reward models from human preferences, which may then be bottlenecked by human performance level, and secondly these reward models require additional human preferences data to further improve.In this work, we study Self-Rewarding Language Models, where the language model itself is used via LLM-as-a-Judge prompting to provide its own rewards during training. We show that during Iterative DPO training, not only does instruction following ability improve, but also the ability to provide high-quality rewards to itself. Fine-tuning Llama 2 70B on three iterations of our approach yields a model that outperforms many existing systems on the AlpacaEval 2.0 leaderboard, including Claude 2, Gemini Pro, and GPT-4 0613. While there is much left still to explore, this work opens the door to the possibility of models that can continually improve in both axes.
|
https://proceedings.mlr.press/v235/yuan24e.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yuan24e/yuan24e.pdf
|
https://openreview.net/forum?id=b89JtZj9gm
|
Not Just Pretty Pictures: Toward Interventional Data Augmentation Using Text-to-Image Generators
|
https://proceedings.mlr.press/v235/yuan24e.html
|
Jianhao Yuan, Francesco Pinto, Adam Davies, Philip Torr
|
https://proceedings.mlr.press/v235/yuan24e.html
|
ICML 2024
|
Neural image classifiers are known to undergo severe performance degradation when exposed to inputs that are sampled from environmental conditions that differ from their training data. Given the recent progress in Text-to-Image (T2I) generation, a natural question is how modern T2I generators can be used to simulate arbitrary interventions over such environmental factors in order to augment training data and improve the robustness of downstream classifiers. We experiment across a diverse collection of benchmarks in single domain generalization (SDG) and reducing reliance on spurious features (RRSF), ablating across key dimensions of T2I generation, including interventional prompting strategies, conditioning mechanisms, and post-hoc filtering, showing that modern T2I generators like Stable Diffusion can indeed be used to implement a powerful interventional data augmentation (IDA) mechanism, outperforming previously state-of-the-art data augmentation techniques regardless of how each dimension is configured.
|
https://proceedings.mlr.press/v235/yuan24f.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yuan24f/yuan24f.pdf
|
https://openreview.net/forum?id=QAGRPiC3FS
|
RigorLLM: Resilient Guardrails for Large Language Models against Undesired Content
|
https://proceedings.mlr.press/v235/yuan24f.html
|
Zhuowen Yuan, Zidi Xiong, Yi Zeng, Ning Yu, Ruoxi Jia, Dawn Song, Bo Li
|
https://proceedings.mlr.press/v235/yuan24f.html
|
ICML 2024
|
Recent advancements in Large Language Models (LLMs) have showcased remarkable capabilities across various tasks in different domains. However, the emergence of biases and the potential for generating harmful content in LLMs, particularly under malicious inputs, pose significant challenges. Current mitigation strategies, while effective, are not resilient under adversarial attacks. This paper introduces Resilient Guardrails for Large Language Models (RigorLLM), a novel framework designed to efficiently and effectively moderate harmful and unsafe inputs and outputs for LLMs. By employing a multi-faceted approach that includes energy-based training data augmentation through Langevin dynamics, optimizing a safe suffix for inputs via minimax optimization, and integrating a fusion-based model combining robust KNN with LLMs based on our data augmentation, RigorLLM offers a robust solution to harmful content moderation. Our experimental evaluations demonstrate that RigorLLM not only outperforms existing baselines like OpenAI API and Perspective API in detecting harmful content but also exhibits unparalleled resilience to jailbreaking attacks. The innovative use of constrained optimization and a fusion-based guardrail approach represents a significant step forward in developing more secure and reliable LLMs, setting a new standard for content moderation frameworks in the face of evolving digital threats.
|
https://proceedings.mlr.press/v235/yue24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yue24a/yue24a.pdf
|
https://openreview.net/forum?id=eG42XBhV9a
|
OLLIE: Imitation Learning from Offline Pretraining to Online Finetuning
|
https://proceedings.mlr.press/v235/yue24a.html
|
Sheng Yue, Xingyuan Hua, Ju Ren, Sen Lin, Junshan Zhang, Yaoxue Zhang
|
https://proceedings.mlr.press/v235/yue24a.html
|
ICML 2024
|
In this paper, we study offline-to-online Imitation Learning (IL) that pretrains an imitation policy from static demonstration data, followed by fast finetuning with minimal environmental interaction. We find the naive combination of existing offline IL and online IL methods tends to behave poorly in this context, because the initial discriminator (often used in online IL) operates randomly and discordantly against the policy initialization, leading to misguided policy optimization and unlearning of pretraining knowledge. To overcome this challenge, we propose a principled offline-to-online IL method, named OLLIE, that simultaneously learns a near-expert policy initialization along with an aligned discriminator initialization, which can be seamlessly integrated into online IL, achieving smooth and fast finetuning. Empirically, OLLIE consistently and significantly outperforms the baseline methods in 20 challenging tasks, from continuous control to vision-based domains, in terms of performance, demonstration efficiency, and convergence speed. This work may serve as a foundation for further exploration of pretraining and finetuning in the context of IL.
|
https://proceedings.mlr.press/v235/yue24b.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yue24b/yue24b.pdf
|
https://openreview.net/forum?id=ZxDqSBgFSM
|
Federated Self-Explaining GNNs with Anti-shortcut Augmentations
|
https://proceedings.mlr.press/v235/yue24b.html
|
Linan Yue, Qi Liu, Weibo Gao, Ye Liu, Kai Zhang, Yichao Du, Li Wang, Fangzhou Yao
|
https://proceedings.mlr.press/v235/yue24b.html
|
ICML 2024
|
Graph Neural Networks (GNNs) have demonstrated remarkable performance in graph classification tasks. However, ensuring the explainability of their predictions remains a challenge. To address this, graph rationalization methods have been introduced to generate concise subsets of the original graph, known as rationales, which serve to explain the predictions made by GNNs. Existing rationalizations often rely on shortcuts in data for prediction and rationale composition. In response, de-shortcut rationalization methods have been proposed, which commonly leverage counterfactual augmentation to enhance data diversity for mitigating the shortcut problem. Nevertheless, these methods have predominantly focused on centralized datasets and have not been extensively explored in the Federated Learning (FL) scenarios. To this end, in this paper, we propose a Federated Graph Rationalization (FedGR) with anti-shortcut augmentations to achieve self-explaining GNNs, which involves two data augmenters. These augmenters are employed to produce client-specific shortcut conflicted samples at each client, which contributes to mitigating the shortcut problem under the FL scenarios. Experiments on real-world benchmarks and synthetic datasets validate the effectiveness of FedGR under the FL scenarios.
|
https://proceedings.mlr.press/v235/yue24c.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yue24c/yue24c.pdf
|
https://openreview.net/forum?id=oOlooUu2Sb
|
How to Leverage Diverse Demonstrations in Offline Imitation Learning
|
https://proceedings.mlr.press/v235/yue24c.html
|
Sheng Yue, Jiani Liu, Xingyuan Hua, Ju Ren, Sen Lin, Junshan Zhang, Yaoxue Zhang
|
https://proceedings.mlr.press/v235/yue24c.html
|
ICML 2024
|
Offline Imitation Learning (IL) with imperfect demonstrations has garnered increasing attention owing to the scarcity of expert data in many real-world domains. A fundamental problem in this scenario is how to extract positive behaviors from noisy data. In general, current approaches to the problem select data building on state-action similarity to given expert demonstrations, neglecting precious information in (potentially abundant) diverse state-actions that deviate from expert ones. In this paper, we introduce a simple yet effective data selection method that identifies positive behaviors based on their resultant states - a more informative criterion enabling explicit utilization of dynamics information and effective extraction of both expert and beneficial diverse behaviors. Further, we devise a lightweight behavior cloning algorithm capable of leveraging the expert and selected data correctly. In the experiments, we evaluate our method on a suite of complex and high-dimensional offline IL benchmarks, including continuous-control and vision-based tasks. The results demonstrate that our method achieves state-of-the-art performance, outperforming existing methods on 20/21 benchmarks, typically by 2-5x, while maintaining a comparable runtime to Behavior Cloning (BC).
|
https://proceedings.mlr.press/v235/yue24d.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/yue24d/yue24d.pdf
|
https://openreview.net/forum?id=oDUJmNCV8D
|
Image Restoration Through Generalized Ornstein-Uhlenbeck Bridge
|
https://proceedings.mlr.press/v235/yue24d.html
|
Conghan Yue, Zhengwei Peng, Junlong Ma, Shiyan Du, Pengxu Wei, Dongyu Zhang
|
https://proceedings.mlr.press/v235/yue24d.html
|
ICML 2024
|
Diffusion models exhibit powerful generative capabilities enabling noise mapping to data via reverse stochastic differential equations. However, in image restoration, the focus is on the mapping relationship from low-quality to high-quality images. Regarding this issue, we introduce the Generalized Ornstein-Uhlenbeck Bridge (GOUB) model. By leveraging the natural mean-reverting property of the generalized OU process and further eliminating the variance of its steady-state distribution through the Doob’s h–transform, we achieve diffusion mappings from point to point enabling the recovery of high-quality images from low-quality ones. Moreover, we unravel the fundamental mathematical essence shared by various bridge models, all of which are special instances of GOUB and empirically demonstrate the optimality of our proposed models. Additionally, we present the corresponding Mean-ODE model adept at capturing both pixel-level details and structural perceptions. Experimental outcomes showcase the state-of-the-art performance achieved by both models across diverse tasks, including inpainting, deraining, and super-resolution. Code is available at https://github.com/Hammour-steak/GOUB.
|
https://proceedings.mlr.press/v235/zaher24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/zaher24a/zaher24a.pdf
|
https://openreview.net/forum?id=qoOt02l2WC
|
Manifold Integrated Gradients: Riemannian Geometry for Feature Attribution
|
https://proceedings.mlr.press/v235/zaher24a.html
|
Eslam Zaher, Maciej Trzaskowski, Quan Nguyen, Fred Roosta
|
https://proceedings.mlr.press/v235/zaher24a.html
|
ICML 2024
|
In this paper, we dive into the reliability concerns of Integrated Gradients (IG), a prevalent feature attribution method for black-box deep learning models. We particularly address two predominant challenges associated with IG: the generation of noisy feature visualizations for vision models and the vulnerability to adversarial attributional attacks. Our approach involves an adaptation of path-based feature attribution, aligning the path of attribution more closely to the intrinsic geometry of the data manifold. Our experiments utilise deep generative models applied to several real-world image datasets. They demonstrate that IG along the geodesics conforms to the curved geometry of the Riemannian data manifold, generating more perceptually intuitive explanations and, subsequently, substantially increasing robustness to targeted attributional attacks.
|
https://proceedings.mlr.press/v235/zahid24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/zahid24a/zahid24a.pdf
|
https://openreview.net/forum?id=6VQXLUy4sQ
|
Sample as you Infer: Predictive Coding with Langevin Dynamics
|
https://proceedings.mlr.press/v235/zahid24a.html
|
Umais Zahid, Qinghai Guo, Zafeirios Fountas
|
https://proceedings.mlr.press/v235/zahid24a.html
|
ICML 2024
|
We present Langevin Predictive Coding (LPC), a novel algorithm for deep generative model learning that builds upon the predictive coding framework of computational neuroscience. By injecting Gaussian noise into the predictive coding inference procedure and incorporating an encoder network initialization, we reframe the approach as an amortized Langevin sampling method for optimizing a tight variational lower bound. To increase robustness to sampling step size, we present a lightweight preconditioning technique inspired by Riemannian Langevin methods and adaptive SGD. We compare LPC against VAEs by training generative models on benchmark datasets; our experiments demonstrate superior sample quality and faster convergence for LPC in a fraction of SGD training iterations, while matching or exceeding VAE performance across key metrics like FID, diversity and coverage.
|
https://proceedings.mlr.press/v235/zakerinia24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/zakerinia24a/zakerinia24a.pdf
|
https://openreview.net/forum?id=pmsPKIBAu6
|
More Flexible PAC-Bayesian Meta-Learning by Learning Learning Algorithms
|
https://proceedings.mlr.press/v235/zakerinia24a.html
|
Hossein Zakerinia, Amin Behjati, Christoph H. Lampert
|
https://proceedings.mlr.press/v235/zakerinia24a.html
|
ICML 2024
|
We introduce a new framework for studying meta-learning methods using PAC-Bayesian theory. Its main advantage over previous work is that it allows for more flexibility in how the transfer of knowledge between tasks is realized. For previous approaches, this could only happen indirectly, by means of learning prior distributions over models. In contrast, the new generalization bounds that we prove express the process of meta-learning much more directly as learning the learning algorithm that should be used for future tasks. The flexibility of our framework makes it suitable to analyze a wide range of meta-learning mechanisms and even design new mechanisms. Other than our theoretical contributions we also show empirically that our framework improves the prediction quality in practical meta-learning mechanisms.
|
https://proceedings.mlr.press/v235/zamboni24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/zamboni24a/zamboni24a.pdf
|
https://openreview.net/forum?id=LbcNAIgNnB
|
How to Explore with Belief: State Entropy Maximization in POMDPs
|
https://proceedings.mlr.press/v235/zamboni24a.html
|
Riccardo Zamboni, Duilio Cirino, Marcello Restelli, Mirco Mutti
|
https://proceedings.mlr.press/v235/zamboni24a.html
|
ICML 2024
|
Recent works have studied state entropy maximization in reinforcement learning, in which the agent’s objective is to learn a policy inducing high entropy over states visitation (Hazan et al., 2019). They typically assume full observability of the state of the system, so that the entropy of the observations is maximized. In practice, the agent may only get partial observations, e.g., a robot perceiving the state of a physical space through proximity sensors and cameras. A significant mismatch between the entropy over observations and true states of the system can arise in those settings. In this paper, we address the problem of entropy maximization over the true states with a decision policy conditioned on partial observations only. The latter is a generalization of POMDPs, which is intractable in general. We develop a memory and computationally efficient policy gradient method to address a first-order relaxation of the objective defined on belief states, providing various formal characterizations of approximation gaps, the optimization landscape, and the hallucination problem. This paper aims to generalize state entropy maximization to more realistic domains that meet the challenges of applications.
|
https://proceedings.mlr.press/v235/zamfir24a.html
|
https://raw.githubusercontent.com/mlresearch/v235/main/assets/zamfir24a/zamfir24a.pdf
|
https://openreview.net/forum?id=0JXGusc7E2
|
See More Details: Efficient Image Super-Resolution by Experts Mining
|
https://proceedings.mlr.press/v235/zamfir24a.html
|
Eduard Zamfir, Zongwei Wu, Nancy Mehta, Yulun Zhang, Radu Timofte
|
https://proceedings.mlr.press/v235/zamfir24a.html
|
ICML 2024
|
Reconstructing high-resolution (HR) images from low-resolution (LR) inputs poses a significant challenge in image super-resolution (SR). While recent approaches have demonstrated the efficacy of intricate operations customized for various objectives, the straightforward stacking of these disparate operations can result in a substantial computational burden, hampering their practical utility. In response, we introduce SeemoRe, an efficient SR model employing expert mining. Our approach strategically incorporates experts at different levels, adopting a collaborative methodology. At the macro scale, our experts address rank-wise and spatial-wise informative features, providing a holistic understanding. Subsequently, the model delves into the subtleties of rank choice by leveraging a mixture of low-rank experts. By tapping into experts specialized in distinct key factors crucial for accurate SR, our model excels in uncovering intricate intra-feature details. This collaborative approach is reminiscent of the concept of “see more", allowing our model to achieve an optimal performance with minimal computational costs in efficient settings.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.