详细信息
A MADRL-Based Credit Allocation Approach for Interactive Multi-Agents ( SCI-EXPANDED收录 EI收录)
文献类型:期刊文献
英文题名:A MADRL-Based Credit Allocation Approach for Interactive Multi-Agents
作者:Wang, Ershen[1,2,3];Wu, Xiaotong[1];Hong, Chen[4,5];Shang, Xinna[4,5];Wu, Peifeng[6];He, Chenglong[3,7];Qu, Pingping[1]
第一作者:Wang, Ershen
通讯作者:Hong, C[1];Hong, C[2]
机构:[1]Shenyang Aerosp Univ, Sch Elect & Informat Engn, Shenyang 110136, Peoples R China;[2]Shenyang Aerosp Univ, Sch Civil Aviat, Shenyang 110136, Peoples R China;[3]54th Res Inst CETC, Shijiazhuang 050001, Peoples R China;[4]Beijing Union Univ, Multiagent Syst Res Ctr, Beijing 100101, Peoples R China;[5]Beijing Union Univ, Coll Robot, Beijing 100101, Peoples R China;[6]Beijing Union Univ, Sch Urban Rail Transit & Logist, Beijing 100101, Peoples R China;[7]54th Res Inst CETC, State Key Lab Satellite Nav Syst & Equipment Techn, Shijiazhuang 050001, Peoples R China
第一机构:Shenyang Aerosp Univ, Sch Elect & Informat Engn, Shenyang 110136, Peoples R China
通讯机构:[1]corresponding author), Beijing Union Univ, Multiagent Syst Res Ctr, Beijing 100101, Peoples R China;[2]corresponding author), Beijing Union Univ, Coll Robot, Beijing 100101, Peoples R China.|[1141739]北京联合大学机器人学院;[11417]北京联合大学;
年份:2025
外文期刊名:INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY & DECISION MAKING
收录:;EI(收录号:20251118051215);Scopus(收录号:2-s2.0-86000736768);WOS:【SCI-EXPANDED(收录号:WOS:001443853900001)】;
基金:This work is supported by the National Key R&D Program of China (Grant No. 2018AAA0100804); the National Natural Science Foundation of China (62173237); the Open Fund of State Key Laboratory of Satellite Navigation System and Equipment Technology (CEPNT2022A01, CEPNT2022B04); the SongShan Laboratory Foundation, China (No. YYJC062022017); Liaoning Provincial Universities' Fundamental Research Funds for Basic Scientific Research (Grant No. Z20240177).
语种:英文
外文关键词:Reinforcement learning; deep neural network; multi-agent; credit allocation
摘要:In multi-agent systems (MAS), the interactions and credit allocation among agents are essential for achieving efficient cooperation. To enhance the interactivity and efficiency of credit allocation in multi-agent reinforcement learning, we introduce a credit allocation for interactive multi-agents method (CAIM). CAIM not only considers the effects of various actions on other agents but also leverages attention mechanisms to handle the mismatch between observations and actions. With a unique credit allocation strategy, agents can more precisely assess their contributions during collaboration. Experiments in various adversarial scenarios within the SMAC benchmark environment indicate that CAIM markedly outperforms existing multi-agent reinforcement learning approaches. Further ablation studies confirm the effectiveness of each CAIM component. This research presents a new paradigm for enhancing collaboration efficiency and overall performance in MAS.
参考文献:
正在载入数据...