摘要
使用自适应遗传算法解决中国象棋计算机博弈问题.将博弈问题分解为搜索引擎、走法生成、评估函数和开局库四大模块,然后将自适应遗传算法引入到评估函数中,通过锦标赛算法对评估函数中的参数组合进行自动调整和优化.设计并开发了基于上述方法的离线自学习系统,实验结果证明提高了程序的棋力.
Adaptive Genetic Algorithm (AGA) is used to ,solve the problem of chinese chess game on computer. The game system is divided into four parts: searching engine, move generator, evaluation function, and opening book. Then, the AGA is brought into evaluation function to automatically adjust and optimize the parameter' combination of evaluation function by tournament algorithm. Based on the above methods, an off-time self-study system is designed. The experiment results show that the AGA improves the performance of the program developed.
出处
《东北大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2005年第10期949-952,共4页
Journal of Northeastern University(Natural Science)
基金
国家自然科学基金资助项目(60475036)
教育部博士点基金资助项目(20040145012)
关键词
中国象棋计算机博弈
博弈树
评估函数
锦标赛算法
自适应遗传算法
Chinese chess computer game
game tree
evaluation function
tournament algorithm
adaptive genetic algorithm