.png){kind=icon}
返回
Mixed-Strategy Learning With Continuous Action Sets
.png){kind=icon}
OA
AI
DOI：10.1109/TAC.2015.2511930.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Motivated by the recent applications of game-theoretical learning to the design of distributed control systems, we study a class of control problems that can be formulated as potential games with continuous action sets. We propose an actor-critic reinforcement learning algorithm that adapts mixed strategies over continuous action spaces. To analyze the algorithm, we extend the theory of finite-dimensional two-timescale stochastic approximation to a Banach space setting, and prove that the continuous dynamics of the process converge to equilibrium in the case of potential games. These results combine to give a provably-convergent learning algorithm in which players do not need to keep track of the controls selected by other agents.
Keyword:
Control systems: Decentralised control
Mathematics: Algorithms: Distributed algorithms
Mathematics: Functional analysis
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
7
论文数:
1.3W
被引数:
6.7W
机构
引用论文
分享收藏Feeding Habits of the Pond Frog, Rana nigromaculata, Inhabiting Rice Fields in Kyoto, Japan.png)
Copeia
CopeiaIF0
PREAI
分享收藏Immunogenicity Testing of Therapeutic Antibodies in Ocular Fluids after Intravitreal InjectionBioanalysisOAAI
BioanalysisIF0
OAAI分享收藏
查看更多内容