DelAC: A Multi-agent Reinforcement Learning of Team-Symmetric Stochastic Games

TL;DR

This paper introduces a multi-agent reinforcement learning algorithm for team-symmetric stochastic games, demonstrating its superior performance through simulations and providing theoretical insights into Nash equilibria.

Contribution

It develops a novel actor-critic algorithm tailored for team-symmetric games and offers a linear complementarity approach to find Nash equilibria.

Findings

The algorithm outperforms existing methods in simulations.

Team-symmetric games always have a Nash equilibrium.

A linear complementarity problem characterizes team-symmetric Nash equilibria.

Abstract

In this paper we study team-symmetric games with $m\ge 2$ teams. Players within a team have symmetric identity and have a common payoff function. We show that team-symmetric games always have a team-symmetric Nash equilibrium. We develop and solve a linear complementarity problem of team-symmetric Nash equilibria. We propose an actor-critic based multi-agent reinforcement learning algorithm for team-symmetric games. Through simulations, we show that this multi-agent reinforcement learning algorithm performs much better than many existing algorithms.