Deep Coordination Graphs

Wendelin Boehmer, Vitaly Kurin, Shimon Whiteson

TL;DR: We introduce an efficient value factorization architecture for MARL that is defined by a coordination graph.

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 factorizing 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 and parameter sharing improves generalization over the state-action space. We show that DCG can solve challenging predator-prey tasks that are vulnerable to the relative overgeneralization pathology and in which all other known value factorization approaches fail.

Keywords: multi-agent reinforcement learning, coordination graph, deep Q-learning, value factorization, relative overgeneralization

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:1910.00091/code)