中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
Dual Self-Awareness Value Decomposition Framework without Individual Global Max for Cooperative MARL

文献类型:会议论文

作者Zhiwei Xu1,2; Bin Zhang1,2; Dapeng Li1,2; Guangchong Zhou1,2; Zeren Zhang1,2; Guoliang Fan1,2
出版日期2023
会议日期December 10-16, 2023
会议地点New Orleans, LA, USA
英文摘要

Value decomposition methods have gained popularity in the field of cooperative multi-agent reinforcement learning. However, almost all existing methods follow the principle of Individual Global Max (IGM) or its variants, which limits their problem-solving capabilities. To address this, we propose a dual self-awareness value decomposition framework, inspired by the notion of dual self-awareness in psychology, that entirely rejects the IGM premise. Each agent consists of an ego policy for action selection and an alter ego value function to solve the credit assignment problem. The value function factorization can ignore the IGM assumption by utilizing an explicit search procedure. On the basis of the above, we also suggest a novel anti-ego exploration mechanism to avoid the algorithm becoming stuck in a local optimum. As the first fully IGM-free value decomposition method, our proposed framework achieves desirable performance in various cooperative tasks.

语种英语
URL标识查看原文
源URL[http://ir.ia.ac.cn/handle/173211/56538]  
专题融合创新中心_决策指挥与体系智能
通讯作者Guoliang Fan
作者单位1.Institute of Automation, Chinese Academy of Sciences
2.School of Artificial Intelligence, University of Chinese Academy of Sciences
推荐引用方式
GB/T 7714
Zhiwei Xu,Bin Zhang,Dapeng Li,et al. Dual Self-Awareness Value Decomposition Framework without Individual Global Max for Cooperative MARL[C]. 见:. New Orleans, LA, USA. December 10-16, 2023.

入库方式: OAI收割

来源:自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。