Pareto coevolution: Using performance against coevolved opponents in a game as dimensions for Pareto selection

摘要

When using an automatic discovery method to find a good strategy in a game, we hope to find one that performs well against a wide variety of opponents. An appealing notion in the use of evolutionary algorithms to coevolve strategies is that the population represents a set of different strategies against which a player must do well. Implicit here is the idea that different players represent different "dimensions" of the domain, and being a robust player means being good in many (preferably all) dimensions of the game. Pareto coevolution makes this idea of "players as dimensions" explicit. By explicitly treating each player as a dimension, or objective, we may then use established multi-objective optimization techniques to find robust strategies. In this paper, we apply Pareto coevolution to Texas Hold'em poker, a complex real-world game of imperfect information. The performance of our Pareto coevolution algorithm is compared with that of a conventional genetic algorithm and shown to be promising.