Clique-based cooperative multiagent reinforcement learning using factor graphs

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 深度强化学习

	Clique-based cooperative multiagent reinforcement learning using factor graphs
	Zhang,Zhen1 ; Zhao DB(赵冬斌)2
发表期刊	IEEE/CAA Journal of Automatica Sinica
	2015
卷号	3 期号:1 页码:248-256
摘要	In this paper, we propose a clique-based sparse reinforcement learning (RL) algorithm for solving cooperative tasks. The aim is to accelerate the learning speed of the original sparse RL algorithm and to make it applicable for tasks decomposed in a more general manner. First, a transition function is estimated and used to update the Q-value function, which greatly reduces the learning time. Second, it is more reasonable to divide agents into cliques, each of which is only responsible for a specific subtask. In this way, the global Q-value function is decomposed into the sum of several simpler local Q-value functions. Such decomposition is expressed by a factor graph and exploited by the general maxplus algorithm to obtain the greedy joint action. Experimental results show that the proposed approach outperforms others with better performance.
关键词	Reinforcement Learning Factor Graphs
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/19321
专题	多模态人工智能系统全国重点实验室_深度强化学习
作者单位	1.Department of Electric Engineering, College of Automation Engineering, Qingdao University 2.State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences, China
推荐引用方式 GB/T 7714	Zhang,Zhen,Zhao DB. Clique-based cooperative multiagent reinforcement learning using factor graphs[J]. IEEE/CAA Journal of Automatica Sinica,2015,3(1):248-256.
APA	Zhang,Zhen,&Zhao DB.(2015).Clique-based cooperative multiagent reinforcement learning using factor graphs.IEEE/CAA Journal of Automatica Sinica,3(1),248-256.
MLA	Zhang,Zhen,et al."Clique-based cooperative multiagent reinforcement learning using factor graphs".IEEE/CAA Journal of Automatica Sinica 3.1(2015):248-256.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Clique-based Coopera（707KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载