Offline Pre-trained Multi-agent Decision Transformer

doi:10.1007/s11633-022-1383-7

CASIA OpenIR > 学术期刊 > Machine Intelligence Research

	Offline Pre-trained Multi-agent Decision Transformer
	Linghui Meng1,2 ; Muning Wen 3; Chenyang Le 3; Xiyun Li1,4 ; Dengpeng Xing1,2 ; Weinan Zhang 3; Ying Wen 3; Haifeng Zhang 1,2; Jun Wang 6; Yaodong Yang 5; Bo Xu1,2
发表期刊	Machine Intelligence Research
ISSN	2731-538X
	2023
卷号	20 期号:2 页码:233-248
摘要	Offline reinforcement learning leverages previously collected offline datasets to learn optimal policies with no necessity to access the real environment. Such a paradigm is also desirable for multi-agent reinforcement learning (MARL) tasks, given the combinatorially increased interactions among agents and with the environment. However, in MARL, the paradigm of offline pre-training with online fine-tuning has not been studied, nor even datasets or benchmarks for offline MARL research are available. In this paper, we facilitate the research by providing large-scale datasets and using them to examine the usage of the decision transformer in the context of MARL. We investigate the generalization of MARL offline pre-training in the following three aspects: 1) between single agents and multiple agents, 2) from offline pretraining to online fine tuning, and 3) to that of multiple downstream tasks with few-shot and zero-shot capabilities. We start by introducing the first offline MARL dataset with diverse quality levels based on the StarCraftII environment, and then propose the novel architecture of multi-agent decision transformer (MADT) for effective offline learning. MADT leverages the transformer′s modelling ability for sequence modelling and integrates it seamlessly with both offline and online MARL tasks. A significant benefit of MADT is that it learns generalizable policies that can transfer between different types of agents under different task scenarios. On the StarCraft II offline dataset, MADT outperforms the state-of-the-art offline reinforcement learning (RL) baselines, including BCQ and CQL. When applied to online tasks, the pre-trained MADT significantly improves sample efficiency and enjoys strong performance in both few-short and zero-shot cases. To the best of our knowledge, this is the first work that studies and demonstrates the effectiveness of offline pre-trained models in terms of sample efficiency and generalizability enhancements for MARL.
关键词	Pre-training model multi-agent reinforcement learning (MARL) decision making transformer offline reinforcement learning
DOI	10.1007/s11633-022-1383-7
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/55977
专题	学术期刊_Machine Intelligence Research
作者单位	1.Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China 2.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China 3.Shanghai Jiao Tong University, Shanghai 200240, China 4.School of Future Technology, University of Chinese Academy of Sciences, Beijing 100049, China 5.Institute for AI, Peking University, Beijing 100871, China 6.Department of Computer Science, University College London, London WC1E 6BT, UK
第一作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Linghui Meng,Muning Wen,Chenyang Le,et al. Offline Pre-trained Multi-agent Decision Transformer[J]. Machine Intelligence Research,2023,20(2):233-248.
APA	Linghui Meng.,Muning Wen.,Chenyang Le.,Xiyun Li.,Dengpeng Xing.,...&Bo Xu.(2023).Offline Pre-trained Multi-agent Decision Transformer.Machine Intelligence Research,20(2),233-248.
MLA	Linghui Meng,et al."Offline Pre-trained Multi-agent Decision Transformer".Machine Intelligence Research 20.2(2023):233-248.

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
MIR-2022-07-219.pdf（2121KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览下载