Learning Efficient Correlated Equilibria

TL;DR

This paper introduces a distributed learning algorithm that guarantees convergence to efficient correlated equilibria by incorporating a common random signal, enabling more optimal collective behavior than traditional Nash equilibrium-focused methods.

Contribution

The paper presents the first distributed learning algorithm that converges to efficient correlated equilibria using a common random signal.

Findings

Algorithm guarantees convergence to efficient correlated equilibria

Incorporating common random signals enhances collective efficiency

High probability of achieving the desired equilibrium

Abstract

The majority of distributed learning literature focuses on convergence to Nash equilibria. Correlated equilibria, on the other hand, can often characterize more efficient collective behavior than even the best Nash equilibrium. However, there are no existing distributed learning algorithms that converge to specific correlated equilibria. In this paper, we provide one such algorithm which guarantees that the agents' collective joint strategy will constitute an efficient correlated equilibrium with high probability. The key to attaining efficient correlated behavior through distributed learning involves incorporating a common random signal into the learning environment.