paper-with-me

홈 › Papers

DMC-VB: A Benchmark for Representation Learning for Control with Visual Distractors

2024-09-26 · Joseph Ortiz, Antoine Dedieu, Wolfgang Lehrach, Swaroop Guntupalli, Carter Wendelken, Ahmad Humayun, Guangyao Zhou, Sivaramakrishnan Swaminathan, Miguel Lázaro-Gredilla, Kevin Murphy

Learning from previously collected data via behavioral cloning or offline reinforcement learning (RL) is a powerful recipe for scaling generalist agents by avoiding the need for expensive online learning. Despite strong generalization in some respects, agents are often remarkably brittle to minor visual variations in control-irrelevant factors such as the background or camera viewpoint. In this paper, we present theDeepMind Control Visual Benchmark (DMC-VB), a dataset collected in the DeepMind Control Suite to evaluate the robustness of offline RL agents for solving continuous control tasks from visual input in the presence of visual distractors. In contrast to prior works, our dataset (a) combines locomotion and navigation tasks of varying difficulties, (b) includes static and dynamic visual variations, (c) considers data generated by policies with different skill levels, (d) systematically returns pairs of state and pixel observation, (e) is an order of magnitude larger, and (f) includes tasks with hidden goals. Accompanying our dataset, we propose three benchmarks to evaluate representation learning methods for pretraining, and carry out experiments on several recently proposed methods. First, we find that pretrained representations do not help policy learning on DMC-VB, and we highlight a large representation gap between policies learned on pixel observations and on states. Second, we demonstrate when expert data is limited, policy learning can benefit from representations pretrained on (a) suboptimal data, and (b) tasks with stochastic hidden goals. Our dataset and benchmark code to train and evaluate agents are available at: https://github.com/google-deepmind/dmc_vision_benchmark.

📄 PDF Abstract BibTeX arXiv:2409.18330

Code (1)

google-deepmind/dmc_vision_benchmark 공식 구현 jax

Tasks

continuous-controlContinuous ControlOffline RLReinforcement Learning (RL)Representation Learning

Similar Papers 제목 키워드 기반

AD3: Implicit Action is the Key for World Models to Distinguish the Diverse Visual Distractors

2024-03-15 · Yucen Wang, Shenghua Wan, Le Gan, Shuai Feng 외

Model-based methods have significantly contributed to distinguishing task-irrelevant distractors for visual control. However, prior research has primarily focused on heterogeneous distractors like noisy background videos…

Understanding What Affects the Generalization Gap in Visual Reinforcement Learning: Theory and Empirical Evidence

2024-02-05 · Jiafei Lyu, Le Wan, Xiu Li, Zongqing Lu

Recently, there are many efforts attempting to learn useful policies for continuous control in visual reinforcement learning (RL). In this scenario, it is important to learn a generalizable policy, as the testing environ…

continuous-controlContinuous ControlLearning TheoryReinforcement Learning (RL)

Task-Induced Representation Learning

2022-04-25 · ICLR 2022 4 · Jun Yamada, Karl Pertsch, Anisha Gunjal, Joseph J. Lim

In this work, we evaluate the effectiveness of representation learning approaches for decision making in visually complex environments. Representation learning is essential for effective reinforcement learning (RL) from …

Contrastive LearningDecision MakingImitation LearningReinforcement Learning (RL)+1

Learning Invariant Visual Representations for Planning with Joint-Embedding Predictive World Models

2026-02-20 · Leonardo F. Toso, Davit Shadunts, Yunyang Lu, Nihal Sharma 외 arxiv

World models learned from high-dimensional visual observations allow agents to make decisions and plan directly in latent space, avoiding pixel-level reconstruction. However, recent latent predictive architectures (JEPAs…

Reimagination with Test-time Observation Interventions: Distractor-Robust World Model Predictions for Visual Model Predictive Control

2025-06-19 · Yuxin Chen, Jianglan Wei, Chenfeng Xu, Boyi Li 외

World models enable robots to "imagine" future observations given current observations and planned actions, and have been increasingly adopted as generalized dynamics models to facilitate robot learning. Despite their pr…

modelModel Predictive Control