arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.12622cs.LGcs.AI
收藏

WD3:缓解深度强化学习中的估计偏差

WD3: Taming the Estimation Bias in Deep Reinforcement Learning

Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院)

Qiang He, Xinwen Hou

AI总结:

针对深度强化学习中价值函数估计的高低估偏差问题,提出加权延迟深度确定性策略梯度算法WD3,通过加权双评论家输出消除偏差,在连续控制任务上性能达到或超过当前最优水平。

摘要:

由函数近似导致的高估现象是深度Q网络、DDPG等基于价值的强化学习算法中的已知问题,会导致策略次优。为解决该问题,TD3采用一对评论家网络输出的最小值。本文证明,在温和假设下,TD3算法会引入低估偏差。为获得更精确的价值函数估计,本文统一这两种对立偏差,提出加权延迟深度确定性策略梯度(Weighted Delayed Deep Deterministic Policy Gradient,WD3)算法,通过对一对评论家网络的输出加权,消除估计偏差并进一步提升性能。为验证WD3的有效性,本文对比了DDPG、TD3与WD3的价值函数学习过程,结果证实该算法确实消除了价值函数的估计误差。此外,本文在连续控制任务上评估了算法,发现在所有测试任务中,WD3的性能始终优于或至少持平于当前最优算法。代码可在https://sites.google.com/view/ictai20-wd3/获取。

英文摘要:

The overestimation phenomenon caused by function approximation is a well-known issue in value-based reinforcement learning algorithms such as deep Q-networks and DDPG, which could lead to suboptimal policies. To address this issue, TD3 takes the minimum value between a pair of critics. In this paper, we show that the TD3 algorithm introduces underestimation bias in mild assumptions. To obtain a more precise estimation for value function, we unify these two opposites and propose a novel algorithm \underline{W}eighted \underline{D}elayed \underline{D}eep \underline{D}eterministic Policy Gradient (WD3), which can eliminate the estimation bias and further improve the performance by weighting a pair of critics. To demonstrate the effectiveness of WD3, we compare the learning process of value function between DDPG, TD3, and WD3. The results verify that our algorithm does eliminate the estimation error of value functions. Furthermore, we evaluate our algorithm on the continuous control tasks. We observe that in each test task, the performance of WD3 consistently outperforms, or at the very least matches, that of the state-of-the-art algorithms\footnote{Our code is available at~\href{https://sites.google.com/view/ictai20-wd3/}{https://sites.google.com/view/ictai20-wd3/}.}.

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。

↑