arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.02701cs.LGcs.AIstat.ML
收藏

理解视觉强化学习中影响泛化差距的因素:理论与实证证据

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

Tsinghua Shenzhen International Graduate School(清华大学深圳国际研究生院);Tsinghua University(清华大学);IEG, Tencent(腾讯IEG);School of Computer Science, Peking University(北京大学计算机学院)

Jiafei Lyu, Le Wan, Xiu Li, Zongqing Lu

AI总结:

本文从理论与实证两方面揭示了视觉强化学习中测试环境存在干扰物时影响泛化差距的关键因素,证明最小化训练与测试环境间的表征距离是缩小泛化差距的最核心手段,并在 DMC-GB 基准上验证了该结论。

摘要:

最近,有许多努力试图在视觉强化学习(RL)中学习用于连续控制的有用策略。在这种情况下,学习可泛化的策略很重要,因为测试环境可能与训练环境不同,例如在部署期间存在干扰物。许多实用算法被提出来处理这个问题。然而,据我们所知,它们都没有提供关于什么影响泛化差距以及为什么其提出的方法有效的理论理解。在本文中,我们通过在理论上回答当测试环境存在干扰物时导致泛化差距的关键因素来弥补这一问题。我们的理论表明,最小化训练和测试环境之间的表征距离(这符合人类直觉)对于减少泛化差距的好处是最关键的。我们的理论结果得到了 DMControl Generalization Benchmark (DMC-GB) 中实证证据的支持。

英文摘要:

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 environment may differ from the training environment, e.g., there exist distractors during deployment. Many practical algorithms are proposed to handle this problem. However, to the best of our knowledge, none of them provide a theoretical understanding of what affects the generalization gap and why their proposed methods work. In this paper, we bridge this issue by theoretically answering the key factors that contribute to the generalization gap when the testing environment has distractors. Our theories indicate that minimizing the representation distance between training and testing environments, which aligns with human intuition, is the most critical for the benefit of reducing the generalization gap. Our theoretical results are supported by the empirical evidence in the DMControl Generalization Benchmark (DMC-GB).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑