Improving Generalization of Multi-agent Reinforcement Learning through Domain-Invariant Feature Extraction

	Improving Generalization of Multi-agent Reinforcement Learning through Domain-Invariant Feature Extraction
	Xu YF(徐一凡); Pu ZQ(蒲志强); Cai QA(蔡奇昂); Li FM(李非墨); Chai XH(柴兴华)
	2023-09
会议名称	International Conference on Artificial Neural Networks
会议日期	2023-5
会议地点	Greece
摘要	The limited generalization ability of reinforcement learning constrains its potential applications, particularly in complex scenarios such as multi-agent systems. To overcome this limitation and enhance the generalization capability of MARL algorithms, this paper proposes a three-stage method that integrates domain randomization and domain adaptation to extract effective features for policy learning. Specifically, the first stage samples environments provided for training and testing in the following stages using domain randomization. The second stage pretrains a domain-invariant feature extractor (DIFE) which employs cycle consistency to disentangle domain-invariant and domain-specific features. The third stage utilizes DIFE for policy learning. Experimental results in MPE tasks demonstrate that our approach yields better performance and generalization ability. Meanwhile, the features captured by DIFE are more interpretable for subsequent policy learning in visualization analysis.
收录类别	EI
七大方向——子方向分类	多智能体系统
国重实验室规划方向分类	多智能体决策
是否有论文关联数据集需要存交	否
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/57458
专题	复杂系统认知与决策实验室_飞行器智能技术
作者单位	Institute of Automation, Chinese Academy of Sciences
推荐引用方式 GB/T 7714	Xu YF,Pu ZQ,Cai QA,et al. Improving Generalization of Multi-agent Reinforcement Learning through Domain-Invariant Feature Extraction[C],2023.

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
Improving Generaliza（7610KB）	会议论文		开放获取	CC BY-NC-SA	浏览下载