English

Deep Coordination Graphs

Machine Learning 2020-06-24 v4 Artificial Intelligence Multiagent Systems

Abstract

This paper introduces the deep coordination graph (DCG) for collaborative multi-agent reinforcement learning. DCG strikes a flexible trade-off between representational capacity and generalization by factoring the joint value function of all agents according to a coordination graph into payoffs between pairs of agents. The value can be maximized by local message passing along the graph, which allows training of the value function end-to-end with Q-learning. Payoff functions are approximated with deep neural networks that employ parameter sharing and low-rank approximations to significantly improve sample efficiency. We show that DCG can solve predator-prey tasks that highlight the relative overgeneralization pathology, as well as challenging StarCraft II micromanagement tasks.

Keywords

Cite

@article{arxiv.1910.00091,
  title  = {Deep Coordination Graphs},
  author = {Wendelin Böhmer and Vitaly Kurin and Shimon Whiteson},
  journal= {arXiv preprint arXiv:1910.00091},
  year   = {2020}
}

Comments

Accepted at ICML 2020

R2 v1 2026-06-23T11:30:49.759Z