English

MMD-MIX: Value Function Factorisation with Maximum Mean Discrepancy for Cooperative Multi-Agent Reinforcement Learning

Multiagent Systems 2021-06-23 v1 Artificial Intelligence Machine Learning

Abstract

In the real world, many tasks require multiple agents to cooperate with each other under the condition of local observations. To solve such problems, many multi-agent reinforcement learning methods based on Centralized Training with Decentralized Execution have been proposed. One representative class of work is value decomposition, which decomposes the global joint Q-value QjtQ_\text{jt} into individual Q-values QaQ_a to guide individuals' behaviors, e.g. VDN (Value-Decomposition Networks) and QMIX. However, these baselines often ignore the randomness in the situation. We propose MMD-MIX, a method that combines distributional reinforcement learning and value decomposition to alleviate the above weaknesses. Besides, to improve data sampling efficiency, we were inspired by REM (Random Ensemble Mixture) which is a robust RL algorithm to explicitly introduce randomness into the MMD-MIX. The experiments demonstrate that MMD-MIX outperforms prior baselines in the StarCraft Multi-Agent Challenge (SMAC) environment.

Keywords

Cite

@article{arxiv.2106.11652,
  title  = {MMD-MIX: Value Function Factorisation with Maximum Mean Discrepancy for Cooperative Multi-Agent Reinforcement Learning},
  author = {Zhiwei Xu and Dapeng Li and Yunpeng Bai and Guoliang Fan},
  journal= {arXiv preprint arXiv:2106.11652},
  year   = {2021}
}

Comments

7 pages, 2 figures, 2 tables. Accepted by IJCNN 2021

R2 v1 2026-06-24T03:27:39.901Z