文献类型：期刊文献

英文题名：GoMIC: Enhancing Efficient Collaboration in Multiagent Reinforcement Learning Through Group-Specific Mutual Information

作者：Wang, Jichao[1];Li, Yi[1];Li, Yichun[1];Mao, Shuai[2];Dong, Zhaoyang[3];Tang, Yang[1]

机构：[1]East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai, Peoples R China;[2]Nantong Univ, Dept Elect Engn, Nantong 226019, Peoples R China;[3]City Univ Hong Kong, Dept Elect Engn, Hong Kong 999077, Peoples R China

年份：2025

卷号：17

期号：6

起止页码：1536

外文期刊名：IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS

收录：;EI(收录号：20252218528871);WOS:【SCI-EXPANDED(收录号:WOS:001636881700002)】；

基金：This work was supported in part by the National Natural Science Foundation of China under Grant 62233005 and Grant U2441245; in part by the Global STEM Professorship and JC STEM Lab of Future Energy Systems; in part by National Key Laboratory of Space Intelligent Control under Grant HTKJ2024KL502004; and in part by China Postdoctoral Science Foundation under Grant 2024M750904

语种：英文

外文关键词：Collaboration; Correlation; Training; Mutual information; Overfitting; Heuristic algorithms; Green products; Urban areas; STEM; Remote sensing; Dynamic grouping; multiagent reinforcement learning (MARL); mutual information (MI)

摘要：In cooperative multiagent reinforcement learning (MARL), previous research has predominantly concentrated on augmenting cooperation through the optimization of global behavioral correlations between agents, with mutual information (MI) typically serving as a crucial metric for correlation quantification. The existing approaches aim to enhance the behavioral correlation among agents to foster better cooperation and goal alignment by leveraging MI. However, it has been demonstrated that the cooperative capabilities among agents cannot be enhanced merely by directly increasing their overall behavioral correlations, particularly in environments with multiple subtasks or scenarios requiring dynamic team structures. To tackle this challenge, a MARL algorithm named group-oriented MI collaboration (GoMIC) is designed, which dynamically partitions agents and employs MI within each partition as an enhanced reward. GoMIC mitigates excessive reliance of individual policies on team-related information and fosters agents to acquire policies across varying team compositions. Experimental evaluations across various tasks in multiagent particle environment (MPE), level-based foraging (LBF), and StarCraft II (SC2) demonstrate the superior performance of GoMIC over some existing approaches, indicating its potential to improve collaboration in multiagent systems.