摘要
藏族久棋是国家级非物质文化遗产,其博弈研究不仅能够推动人工智能的发展,也有助于促进藏棋文化的保护与传承。久棋的复杂性高于围棋,其博弈分为3个阶段,每个阶段规则不同;其动作与空间复杂度极大,给低资源高效率博弈算法研究带来挑战。研究梳理了当前久棋博弈研究的主要算法,并分析了现有久棋AI的水平。尽管基于专家知识的算法在实际对战中表现较好,但受到专家知识匮乏的限制;而结合知识与数据的深度强化学习算法,虽然在方法上较为先进,但因硬件资源的限制,AI水平提升受限。此外还分析了现有的久棋线上对弈平台,并探讨了当前博弈研究中存在的问题,提出了未来研究的可能方向。
Tibetan Jiu Chess,as a national intangible cultural heritage,not only promotes the development of artificial intelligence,but also contributes to the protection and spread of culture.Jiu Chess is more complex than Go,with its gameplay divided into three stages,each with different rules.The action and spatial complexity are immense,presenting significant challenges for low-resource cost and high efficient game algorithm research.This paper reviews and analyzes the main algorithms in current Jiu Chess game research and evaluates the performance of existing Jiu Chess AI.Although expert knowledge-based algorithms perform well in actual games,they are still limited by its scarcity.In contrast,deep reinforcement learning algorithms integrating knowledge and data are more advanced in methodology,but are constrained by hardware limitations,hindering AI performance improvement.This paper also analyzes existing Jiu Chess online gaming platforms,explores the challenges in current game research,and proposes some suggestions for future research.
作者
李霞丽
顾旌世
高乔
张皓扬
何非凡
LI Xiali;GU Jingshi;GAO Qiao;ZHANG Haoyang;HE Feifan(Key Laboratory of Ethnic Language Intelligent Analysis and Security Governance of MOE,Minzu University of China,Beijing 100081,China;School of Information Engineering,Minzu University of China,Beijing 100081,China;Digital and Smart Campus Construction Center,Beijing University of Chinese Medicine,Beijing 100029,China;Swinburne College,Shandong University of Science and Technology,Jinan 250031,China)
出处
《重庆理工大学学报(自然科学)》
北大核心
2025年第8期90-96,共7页
Journal of Chongqing University of Technology:Natural Science
基金
国家自然科学基金项目(62276285)。
关键词
计算机博弈
藏族久棋
专家知识
深度强化学习
computer games
Tibetan Jiu Chess
expert knowledge
deep reinforcement learning
作者简介
李霞丽,女,教授,主要从事计算机博弈研究,E-mail:lixiali@muc.edu.cn;通信作者:高乔,女,硕士,实验师,主要从事计算机博弈研究,E-mail:gaoq@bucm.edu.cn。
