摘要
机器博弈被认为是人工智能领域最具挑战性的研究方向之一.国际象棋的计算机博弈已经有了很长的历史,并且经历了一场波澜壮阔的“搏杀”,“深蓝”计算机的胜利也给人类留下了难以忘怀的记忆.中国象棋计算机博弈的难度绝不亚于国际象棋,不仅涉足学者太少,而且参考资料不多.在国际象棋成熟技术的基础上,结合在中国象棋机器博弈方面的多年实践,总结出一套过程建模、状态表示、着法生成、棋局评估、博弈树搜索、开局库与残局库开发、系统测试与参数优化等核心技术要点,最后提出了当前研究的热点与方向.
Computer game is one of the most challenging topics in the field of artificial intelligence. Chess computer game has a long history, and come through tough research. Finally, DeepBluet s victory startled the whole world. Chinese chess computer game is more complex than chess computer game, and the fewer researchers and fewer references lead the lag in the field. Based on a series of technologies of chess computer game and years practice of Chinese chess, a set of schemes and methods are proposed, such as the process modeling, state representation, move generation, evaluation function, searching game tree, opening book, endgame database, system test and parameter optimization, etc. Hot topics and tasks are also proposed at the end.
出处
《小型微型计算机系统》
CSCD
北大核心
2006年第6期961-969,共9页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(60475036)资助.
关键词
人工智能
中国象棋计算机博弈
机器博弈过程建模
着法生成
评估函数
博弈树搜索
artificial intelligence （AI）
Chinese chess computer game
process modeling of computer game
move generation
evaluation function
searching game tree
