Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Believe What You See: Implicit Constraint Approach for Offline Multi-Agent Reinforcement Learning

About

Learning from datasets without interaction with environments (Offline Learning) is an essential step to apply Reinforcement Learning (RL) algorithms in real-world scenarios. However, compared with the single-agent counterpart, offline multi-agent RL introduces more agents with the larger state and action space, which is more challenging but attracts little attention. We demonstrate current offline RL algorithms are ineffective in multi-agent systems due to the accumulated extrapolation error. In this paper, we propose a novel offline RL algorithm, named Implicit Constraint Q-learning (ICQ), which effectively alleviates the extrapolation error by only trusting the state-action pairs given in the dataset for value estimation. Moreover, we extend ICQ to multi-agent tasks by decomposing the joint-policy under the implicit constraint. Experimental results demonstrate that the extrapolation error is successfully controlled within a reasonable range and insensitive to the number of agents. We further show that ICQ achieves the state-of-the-art performance in the challenging multi-agent offline tasks (StarCraft II). Our code is public online at https://github.com/YiqinYang/ICQ.

Yiqin Yang, Xiaoteng Ma, Chenghao Li, Zewu Zheng, Qiyuan Zhang, Gao Huang, Jun Yang, Qianchuan Zhao• 2021

Related benchmarks

TaskDatasetResultRank
Offline Reinforcement LearningD4RL Adroit (expert, human)
Adroit Door Return (Human)6.4
29
Multi-Agent Reinforcement LearningSMAC corridor (test)
Average Score16.74
12
Multi-Agent Reinforcement LearningSMAC 6h_vs_8z (test)
Average Score11.55
12
Offline Reinforcement LearningD4RL AntMaze fixed, play, diverse
AntMaze UMaze (Fixed) Score85
10
StarCraft II micromanagementStarCraft II 2s3z mixed
Win Rate85
8
StarCraft II micromanagementStarCraft II 2s3z medium_replay
Win Rate41
8
StarCraft II micromanagementStarCraft II 5m_vs_6m medium_replay
Win Rate18
8
Multi-agent Offline Reinforcement LearningMPE PP (Medium-replay)
Score34.5
8
StarCraft II micromanagementStarCraft II 2s3z medium
Win Rate18
8
StarCraft II micromanagementStarCraft II 5m_vs_6m medium
Win Rate26
8
Showing 10 of 84 rows
...

Other info

Code

Follow for update