Game Theoretic Rating in N-player general-sum games with Equilibria
This work addresses the challenge of strategy rating in complex multiagent interactions, which is important for applications in game theory, AI, and real-world competitive or cooperative settings, representing an incremental advancement over prior methods.
The paper tackles the problem of rating strategies in N-player general-sum games by proposing novel algorithms that leverage game-theoretic equilibria, enabling efficient strategy rating in complex multiagent settings. The methods are empirically validated on real-world data and multiagent reinforcement learning agent evaluation.
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.