Skip to main content

Deep Coordination Graphs

Wendelin Boehmer‚ Vitaly Kurin and Shimon Whiteson

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.

Book Title
ICML 2020: Proceedings of the Thirty−Seventh International Conference on Machine Learning
Month
July
Year
2020