arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.01779cs.LGcs.AIstat.ML
收藏

近似两个价值函数而不是一个:表征一个新的深度强化学习算法族

Approximating two value functions instead of one: towards characterizing a new family of Deep Reinforcement Learning algorithms

Montefiore Institute(蒙特菲奥雷研究所);Université de Liège(列日大学);University of Groningen(格罗宁根大学)

Matthia Sabatelli, Gilles Louppe, Pierre Geurts, Marco A. Wiering

摘要:

本文朝着表征一个新的\textit{model-free}深度强化学习(DRL)算法家族前进了一步。这些算法的目的是联合学习状态-价值函数($V$)的近似以及状态-动作价值函数($Q$)的近似。我们的分析从深入研究深度质量-价值学习(DQV)算法开始,DRL算法已被证明优于流行的技术,如深度Q学习(DQN)和双深度Q学习(DDQN)\cite{sabatelli2018deep}。为了研究为什么DQV的学习动态允许该算法执行得如此之好,我们制定了一组研究问题,帮助我们表征一个新的家庭的DRL算法。在我们的研究结果中,我们提出了一些具体的情况下,DQV的性能可能会受到损害,并介绍了一种新的\textit{off-policy} DRL算法,称为DQV-Max,它可以优于DQV。然后,我们研究的行为的$V$和$Q$功能,学习DQV和DQV-Max,并表明,这两种算法可能在几个DRL测试台上表现得很好,因为他们不太容易遭受高估偏差的$Q$功能。

英文摘要:

This paper makes one step forward towards characterizing a new family of \textit{model-free} Deep Reinforcement Learning (DRL) algorithms. The aim of these algorithms is to jointly learn an approximation of the state-value function ($V$), alongside an approximation of the state-action value function ($Q$). Our analysis starts with a thorough study of the Deep Quality-Value Learning (DQV) algorithm, a DRL algorithm which has been shown to outperform popular techniques such as Deep-Q-Learning (DQN) and Double-Deep-Q-Learning (DDQN) \cite{sabatelli2018deep}. Intending to investigate why DQV's learning dynamics allow this algorithm to perform so well, we formulate a set of research questions which help us characterize a new family of DRL algorithms. Among our results, we present some specific cases in which DQV's performance can get harmed and introduce a novel \textit{off-policy} DRL algorithm, called DQV-Max, which can outperform DQV. We then study the behavior of the $V$ and $Q$ functions that are learned by DQV and DQV-Max and show that both algorithms might perform so well on several DRL test-beds because they are less prone to suffer from the overestimation bias of the $Q$ function.

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。

↑