Game Theoretic Rating in N-player general-sum games with Equilibria

TL;DR

This paper introduces new algorithms for rating strategies in N-player, general-sum games using game theoretic solutions, enabling more accurate evaluation of strategies in complex multiagent interactions.

Contribution

It generalizes existing rating methods to N-player, general-sum games and proposes algorithms that leverage equilibria for strategy evaluation.

Findings

Validated on real-world data (Premier League)

Effective in multiagent reinforcement learning scenarios

Improves strategy rating accuracy in complex games

Abstract

Rating strategies in a game is an important area of research in game theory and artificial intelligence, and can be applied to any real-world competitive or cooperative setting. Traditionally, only transitive dependencies between strategies have been used to rate strategies (e.g. Elo), however recent work has expanded ratings to utilize game theoretic solutions to better rate strategies in non-transitive games. This work generalizes these ideas and proposes novel algorithms suitable for N-player, general-sum rating of strategies in normal-form games according to the payoff rating system. This enables well-established solution concepts, such as equilibria, to be leveraged to efficiently rate strategies in games with complex strategic interactions, which arise in multiagent training and real-world interactions between many agents. We empirically validate our methods on real world normal-form data (Premier League) and multiagent reinforcement learning agent evaluation.