Efficient Nash Computation in Large Population Games with Bounded Influence

TL;DR

This paper presents a new framework for large-population games with limited player influence, along with efficient algorithms for computing and learning approximate Nash equilibria within this setting.

Contribution

It introduces a generalized game representation with centralized, bounded influence and provides the first provably correct and efficient algorithms for Nash equilibrium computation in this framework.

Findings

Efficient algorithms for approximate Nash equilibria

Generalization of congestion games to broader settings

Provable correctness of the proposed methods

Abstract

We introduce a general representation of large-population games in which each player s influence ON the others IS centralized AND limited, but may otherwise be arbitrary.This representation significantly generalizes the class known AS congestion games IN a natural way.Our main results are provably correct AND efficient algorithms FOR computing AND learning approximate Nash equilibria IN this general framework.