• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
DM$^2$: Decentralized Multi-Agent Reinforcement Learning via Distribution Matching.
Caroline
Wang, Ishan Durugkar, Elad Liebman,
and Peter Stone.
In Proceedings of the 37th AAAI Conference on Artificial
Intelligence (AAAI-23), February 2023.
[PDF]801.2kB [slides.pdf]3.7MB [poster.pdf]1.4MB
Current approaches to multi-agent cooperation rely heavily on centralized mechanisms or explicit communication protocols to ensure convergence. This paper studies the problem of distributed multi-agent learning without resorting to centralized components or explicit communication. It examines the use of distribution matching to facilitate the coordination of independent agents. In the proposed scheme, each agent independently minimizes the distribution mismatch to the corresponding component of a target visitation distribution. The theoretical analysis shows that under certain conditions, each agent minimizing its individual distribution mismatch allows the convergence to the joint policy that generated the target distribution. Further, if the target distribution is from a joint policy that optimizes a cooperative task, the optimal policy for a combination of this task reward and the distribution matching reward is the same joint policy. This insight is used to formulate a practical algorithm (DM$^2$), in which each individual agent matches a target distribution derived from concurrently sampled trajectories from a joint expert policy. Experimental validation on the StarCraft domain shows that combining (1) a task reward, and (2) a distribution matching reward for expert demonstrations for the same task, allows agents to outperform a naive distributed baseline. Additional experiments probe the conditions under which expert demonstrations need to be sampled to obtain the learning benefits.
@InProceedings{aaai23-wang,
author = {Caroline Wang and Ishan Durugkar and Elad Liebman and Peter Stone},
title = {DM$^2$: Decentralized Multi-Agent Reinforcement Learning via Distribution Matching},
booktitle = {Proceedings of the 37th AAAI Conference on Artificial Intelligence (AAAI-23)},
location = {Washington, D.C.},
month = {February},
year = {2023},
abstract = {Current approaches to multi-agent cooperation rely heavily on
centralized mechanisms or explicit communication protocols to ensure
convergence. This paper studies the problem of distributed multi-agent
learning without resorting to centralized components or explicit
communication. It examines the use of distribution matching to facilitate the
coordination of independent agents. In the proposed scheme, each agent
independently minimizes the distribution mismatch to the corresponding
component of a target visitation distribution. The theoretical analysis shows
that under certain conditions, each agent minimizing its individual
distribution mismatch allows the convergence to the joint policy that
generated the target distribution. Further, if the target distribution is
from a joint policy that optimizes a cooperative task, the optimal policy for
a combination of this task reward and the distribution matching reward is
the same joint policy. This insight is used to formulate a practical
algorithm (DM$^2$), in which each individual agent matches a target
distribution derived from concurrently sampled trajectories from a joint
expert policy. Experimental validation on the StarCraft domain shows that
combining (1) a task reward, and (2) a distribution matching reward for
expert demonstrations for the same task, allows agents to outperform a naive
distributed baseline. Additional experiments probe the conditions under which
expert demonstrations need to be sampled to obtain the learning benefits.},
}
Generated by bib2html.pl (written by Patrick Riley ) on Thu Oct 23, 2025 16:14:15