Policy Space Diversity for Non-Transitive Games

TL;DR

This paper introduces a new diversity metric for policy populations in PSRO algorithms, ensuring better approximation to Nash Equilibria in non-transitive games, and demonstrates its effectiveness through empirical results.

Contribution

The paper proposes a novel diversity metric that guarantees improved NE approximation and develops PSD-PSRO, a new PSRO variant with convergence guarantees and superior empirical performance.

Findings

PSD-PSRO produces less exploitable policies.

The new diversity metric improves NE approximation.

Empirical results show enhanced performance across various games.

Abstract

Policy-Space Response Oracles (PSRO) is an influential algorithm framework for approximating a Nash Equilibrium (NE) in multi-agent non-transitive games. Many previous studies have been trying to promote policy diversity in PSRO. A major weakness in existing diversity metrics is that a more diverse (according to their diversity metrics) population does not necessarily mean (as we proved in the paper) a better approximation to a NE. To alleviate this problem, we propose a new diversity metric, the improvement of which guarantees a better approximation to a NE. Meanwhile, we develop a practical and well-justified method to optimize our diversity metric using only state-action samples. By incorporating our diversity regularization into the best response solving in PSRO, we obtain a new PSRO variant, Policy Space Diversity PSRO (PSD-PSRO). We present the convergence property of PSD-PSRO. Empirically, extensive experiments on various games demonstrate that PSD-PSRO is more effective in producing significantly less exploitable policies than state-of-the-art PSRO variants.