Distributed Non-Bayesian Learning for Games with Incomplete Information

TL;DR

This paper introduces a distributed learning method for games with unknown parameters, enabling agents to learn the true parameter and reach a Nash equilibrium through coupled belief and strategy updates.

Contribution

It proposes a novel non-Bayesian distributed learning rule combined with best response dynamics for games with incomplete information.

Findings

Agents' beliefs converge to a common belief.

Strategy profiles converge to a Nash equilibrium.

Beliefs eventually concentrate on the true parameter.

Abstract

We consider distributed learning problem in games with an unknown cost-relevant parameter, and aim to find the Nash equilibrium while learning the true parameter. Inspired by the social learning literature, we propose a distributed non-Bayesian rule to learn the parameter (each agent maintains a belief on the parameter and updates the belief according to the received noisy cost), combined with best response dynamics for strategy update. The difficulty of the analysis lies mainly in the fact that the parameter learning process and strategy update process are coupled. We first prove that agents' beliefs converge to a common belief and the strategy profiles converge to a Nash equilibrium under this common belief. On this basis, we further show that the beliefs eventually concentrate on the true parameter.