Peer Incentive Reinforcement Learning for Cooperative Multiagent Games

doi:10.1109/TG.2022.3196925

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统认知与决策实验室

	Peer Incentive Reinforcement Learning for Cooperative Multiagent Games
	Zhang, Tianle 1,2; Liu, Zhen 1,2; Pu, Zhiqiang 1,2; Yi, Jianqiang 1,2
刊名	IEEE TRANSACTIONS ON GAMES
	2023-12-01
卷号	15 期号:4 页码:623-636
关键词	Cooperative multiagent games intrinsic reward multiagent reinforcement learning (MARL) Starcraft II Micromanagement
ISSN号	2475-1502
DOI	10.1109/TG.2022.3196925
通讯作者	Liu, Zhen(liuzhen@ia.ac.cn)
英文摘要	Social learning, especially social incentives, is extremely important for humans to achieve a high level of coordination. Inspired by this, we introduce this concept into cooperative multiagent reinforcement learning (MARL), to implicitly address the credit assignment problem and promote the interagent direct interactions for cooperations among agents in cooperative multiagent games. In this article, we propose a novel intrinsic reward method with peer incentives (IRPI) based on actor-critic policy gradient. This method can enable agents to incentivize each other for their cooperations through using causal influence among them. Specifically, a novel intrinsic reward mechanism is innovatively designed to empower each agent the ability to give positive or negative rewards to other peer agents' actions through considering the causal influence of the other agents on it. The mechanism is realized by a feedforward neural network through utilizing causal influence between the agents. The causal influence of one agent on another is inferred via counterfactual reasoning using the joint action-value function in MARL. The quality of the influence is assessed via counterfactual reasoning using the individual value function in MARL. Simulations are carried out on two popular multiagent game testbeds: Starcraft II Micromanagement and Multiagent Particle Environments. Simulation results demonstrate that the proposed IRPI can enhance cooperations among the agents to achieve better performance compared with a number of state-of-the-art MARL methods in a variety of cooperative multiagent games.
资助项目	National Key Research and Development Program of China
WOS关键词	STARCRAFT ; LEVEL
WOS研究方向	Computer Science
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:001128375200007
资助机构	National Key Research and Development Program of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/54904]
专题	复杂系统认知与决策实验室
通讯作者	Liu, Zhen
作者单位	1.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Tianle,Liu, Zhen,Pu, Zhiqiang,et al. Peer Incentive Reinforcement Learning for Cooperative Multiagent Games[J]. IEEE TRANSACTIONS ON GAMES,2023,15(4):623-636.
APA	Zhang, Tianle,Liu, Zhen,Pu, Zhiqiang,&Yi, Jianqiang.(2023).Peer Incentive Reinforcement Learning for Cooperative Multiagent Games.IEEE TRANSACTIONS ON GAMES,15(4),623-636.
MLA	Zhang, Tianle,et al."Peer Incentive Reinforcement Learning for Cooperative Multiagent Games".IEEE TRANSACTIONS ON GAMES 15.4(2023):623-636.