arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.05800cs.LG
收藏

强大的策略梯度可防止严重的数据损坏

Robust Policy Gradient against Strong Data Corruption

University of Wisconsin--Madison(威斯康星大学麦迪逊分校);Cornell University(康奈尔大学)

Xuezhou Zhang, Yiding Chen, Xiaojin Zhu, Wen Sun

摘要:

研究了对抗性腐败下的鲁棒强化学习问题。我们的攻击模型假设一个\textit{adaptive}对手可以任意破坏奖励和过渡在每一步的情节,最多$ε$ -分数的学习情节。我们的攻击模型是严格强于那些考虑在以前的作品。我们的第一个结果表明,没有算法可以找到一个更好的比$O(ε)$ -最优策略下,我们的攻击模型。接下来,我们表明,令人惊讶的自然策略梯度(NPG)方法保留了自然的鲁棒性,如果奖励腐败是有界的,可以找到一个$O(\sqrtε)$最优的政策。因此,我们开发了一个过滤的策略梯度(FPG)算法,甚至可以容忍无限的奖励腐败,并可以找到一个$O(ε^{1/4})$最优的政策。我们强调,FPG是第一个可以实现一个有意义的学习保证时,一个恒定的分数的情节被损坏。补充的理论结果,我们表明,神经实现的FPG实现强大的鲁棒学习性能的MuJoCo连续控制基准。

英文摘要:

We study the problem of robust reinforcement learning under adversarial corruption on both rewards and transitions. Our attack model assumes an \textit{adaptive} adversary who can arbitrarily corrupt the reward and transition at every step within an episode, for at most $ε$-fraction of the learning episodes. Our attack model is strictly stronger than those considered in prior works. Our first result shows that no algorithm can find a better than $O(ε)$-optimal policy under our attack model. Next, we show that surprisingly the natural policy gradient (NPG) method retains a natural robustness property if the reward corruption is bounded, and can find an $O(\sqrtε)$-optimal policy. Consequently, we develop a Filtered Policy Gradient (FPG) algorithm that can tolerate even unbounded reward corruption and can find an $O(ε^{1/4})$-optimal policy. We emphasize that FPG is the first that can achieve a meaningful learning guarantee when a constant fraction of episodes are corrupted. Complimentary to the theoretical results, we show that a neural implementation of FPG achieves strong robust learning performance on the MuJoCo continuous control benchmarks.

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。

↑