On self-play computation of equilibrium in poker
This work addresses the challenge of efficient equilibrium computation in poker for AI and game theory researchers, but it is incremental as it compares existing methods on known game variants.
The paper tackled the problem of computing near-equilibrium strategies in simplified poker games by comparing genetic algorithms and counterfactual regret minimization, finding that both algorithms performed well against analytical Nash equilibrium benchmarks, with specific numerical results indicating high accuracy in strategy approximation.
We compare performance of the genetic algorithm and the counterfactual regret minimization algorithm in computing the near-equilibrium strategies in the simplified poker games. We focus on the von Neumann poker and the simplified version of the Texas Hold'Em poker, and test outputs of the considered algorithms against analytical expressions defining the Nash equilibrium strategies. We comment on the performance of the studied algorithms against opponents deviating from equilibrium.