Scalable Reinforcement Learning for Multi-Agent Networked Systems

TL;DR

This paper introduces a scalable reinforcement learning framework for multi-agent networked systems that leverages network structure to find localized policies efficiently, addressing the exponential growth of state-action spaces.

Contribution

The paper proposes a novel Scalable Actor Critic framework that exploits network locality to achieve efficient approximate solutions in large multi-agent RL problems.

Findings

Achieves $O( ho^{ppa})$-approximation of stationary points.

Complexity scales with local neighborhood size.

Applicable to wireless, epidemic, and traffic networks.

Abstract

We study reinforcement learning (RL) in a setting with a network of agents whose states and actions interact in a local manner where the objective is to find localized policies such that the (discounted) global reward is maximized. A fundamental challenge in this setting is that the state-action space size scales exponentially in the number of agents, rendering the problem intractable for large networks. In this paper, we propose a Scalable Actor Critic (SAC) framework that exploits the network structure and finds a localized policy that is an $O(\rho^{\kappa})$-approximation of a stationary point of the objective for some $\rho\in(0,1)$, with complexity that scales with the local state-action space size of the largest $\kappa$-hop neighborhood of the network. We illustrate our model and approach using examples from wireless communication, epidemics and traffic.