Deep Coordination Graphs

Wendelin Boehmer, Vitaly Kurin, Shimon Whiteson
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:980-991, 2020.

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.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-boehmer20a, title = {Deep Coordination Graphs}, author = {Boehmer, Wendelin and Kurin, Vitaly and Whiteson, Shimon}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {980--991}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/boehmer20a/boehmer20a.pdf}, url = {http://proceedings.mlr.press/v119/boehmer20a.html}, 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.} }
Endnote
%0 Conference Paper %T Deep Coordination Graphs %A Wendelin Boehmer %A Vitaly Kurin %A Shimon Whiteson %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-boehmer20a %I PMLR %P 980--991 %U http://proceedings.mlr.press/v119/boehmer20a.html %V 119 %X 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.
APA
Boehmer, W., Kurin, V. & Whiteson, S.. (2020). Deep Coordination Graphs. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:980-991 Available from http://proceedings.mlr.press/v119/boehmer20a.html.

Related Material