Enhancing Multi-agent Coordination via Dual-channel Consensus

doi:10.1007/s11633-023-1464-2

CASIA OpenIR > 学术期刊 > Machine Intelligence Research

	Enhancing Multi-agent Coordination via Dual-channel Consensus
	Qingyang Zhang1,2 ; Kaishen Wang1,3 ; Jingqing Ruan 1,2; Yiming Yang1 ; Dengpeng Xing1,3 ; Bo Xu1,2,3
发表期刊	Machine Intelligence Research
ISSN	2731-538X
	2024
卷号	21 期号:2 页码:349-368
摘要	Successful coordination in multi-agent systems requires agents to achieve consensus. Previous works propose methods through information sharing, such as explicit information sharing via communication protocols or exchanging information implicitly via behavior prediction. However, these methods may fail in the absence of communication channels or due to biased modeling. In this work, we propose to develop dual-channel consensus (DuCC) via contrastive representation learning for fully cooperative multi-agent systems, which does not need explicit communication and avoids biased modeling. DuCC comprises two types of consensus: temporally extended consensus within each agent (inner-agent consensus) and mutual consensus across agents (inter-agent consensus). To achieve DuCC, we design two objectives to learn representations of slow environmental features for inner-agent consensus and to realize cognitive consistency as inter-agent consensus. Our DuCC is highly general and can be flexibly combined with various MARL algorithms. The extensive experiments on StarCraft multi-agent challenge and Google research football demonstrate that our method efficiently reaches consensus and performs superiorly to state-of-the-art MARL algorithms.
关键词	Multi-agent reinforcement learning, contrastive representation learning, consensus, multi-agent cooperation, cognitive consistency
DOI	10.1007/s11633-023-1464-2
引用统计
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/56043
专题	学术期刊_Machine Intelligence Research
作者单位	1.Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China 2.School of Future Technology, University of Chinese Academy of Sciences, Beijing 100049, China 3.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China
第一作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Qingyang Zhang,Kaishen Wang,Jingqing Ruan,et al. Enhancing Multi-agent Coordination via Dual-channel Consensus[J]. Machine Intelligence Research,2024,21(2):349-368.
APA	Qingyang Zhang,Kaishen Wang,Jingqing Ruan,Yiming Yang,Dengpeng Xing,&Bo Xu.(2024).Enhancing Multi-agent Coordination via Dual-channel Consensus.Machine Intelligence Research,21(2),349-368.
MLA	Qingyang Zhang,et al."Enhancing Multi-agent Coordination via Dual-channel Consensus".Machine Intelligence Research 21.2(2024):349-368.

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
MIR-2023-06-096.pdf（4997KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览下载