文献类型：会议论文

英文题名：An Algorithm Combining Hidden States for Monotonic Value Function Factorisation

作者：Wang, Ershen[1]; Wu, Xiaotong[1]; Hong, Chen[2,3]; Chen, Jihao[1]

第一作者：Wang, Ershen

机构：[1] School of Electronic and Information Engineering, Shenyang Aerospace University, Shenyang, 110136, China; [2] Multi-Agent Systems Research Centre, Beijing Union University, Beijing, 100101, China; [3] College of Robotics, Beijing Union University, Beijing, 100101, China

第一机构：School of Electronic and Information Engineering, Shenyang Aerospace University, Shenyang, 110136, China

会议论文集：2023 3rd International Conference on Big Data, Artificial Intelligence and Risk Management, ICBAR 2023

会议日期：November 24, 2023 - November 26, 2023

会议地点：Chengdu, China

语种：英文

外文关键词：Fertilizers - Game theory - Learning algorithms - Multi agent systems - Recurrent neural networks - Software agents

摘要：By using the communication and shared cooperation information between agents to infer the observation states of other agents, partially observable problems in multi-agent systems can be solved by centralized training decentralized execution (CTDE). However, even in CTDE, agents may still get stuck in local optimal solutions. In order to solve the problem, we design a novel multi-agent reinforcement learning (MARL) algorithm, hidden state based QMIX (HSQMIX), which uses the hidden state generated by recurrent neural networks (RNNs) to replace the true state of the agent as the input of the mixing network, and introduces transformer to process the hidden state. Besides, the priority experience replay and temporal difference learning are elaborately integrated into the algorithm. We also evaluate HSQMIX in the StarCraft multi-agent challenge (SMAC) map and compare it with other popular MARL algorithms. Experimental results show that HSQMIX outperforms other algorithms by a large margin. Our work will provide new insights into the multi-agent cooperative game. ? 2023 ACM.