Genetic Algorithms for Mentor-Assisted Evaluation Function Optimization
This addresses the challenge of optimizing evaluation functions in chess AI, offering a method that could be applied to other problems with available mentors, though it is incremental as it builds on existing genetic algorithm techniques.
The paper tackles the problem of reverse engineering evaluation function parameters for computer chess using genetic algorithms, achieving performance on par with top tournament-playing chess programs and outperforming a two-time World Computer Chess Champion by evolving a program with fewer parameters to mimic a superior mentor.
In this paper we demonstrate how genetic algorithms can be used to reverse engineer an evaluation function's parameters for computer chess. Our results show that using an appropriate mentor, we can evolve a program that is on par with top tournament-playing chess programs, outperforming a two-time World Computer Chess Champion. This performance gain is achieved by evolving a program with a smaller number of parameters in its evaluation function to mimic the behavior of a superior mentor which uses a more extensive evaluation function. In principle, our mentor-assisted approach could be used in a wide range of problems for which appropriate mentors are available.