期刊
SIAM JOURNAL ON CONTROL AND OPTIMIZATION
卷 42, 期 4, 页码 1143-1166出版社
SIAM PUBLICATIONS
DOI: 10.1137/S0363012901385691
关键词
reinforcement learning; Markov decision processes; actor-critic algorithms; stochastic approximation
In this article, we propose and analyze a class of actor-critic algorithms. These are two-time-scale algorithms in which the critic uses temporal difference learning with a linearly parameterized approximation architecture, and the actor is updated in an approximate gradient direction, based on information provided by the critic. We show that the features for the critic should ideally span a subspace prescribed by the choice of parameterization of the actor. We study actor-critic algorithms for Markov decision processes with Polish state and action spaces. We state and prove two results regarding their convergence.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据