[1]高强,徐心和,王昊,等.一种基于经验的德州扑克博弈系统架构[J].智能系统学报,2020,15(3):468-474.[doi:10.11992/tis.201803043]
　GAO Qiang,XU Xinhe,WANG Hao,et al.System architecture of Texas Hold’em based on experience[J].CAAI Transactions on Intelligent Systems,2020,15(3):468-474.[doi:10.11992/tis.201803043]
点击复制

一种基于经验的德州扑克博弈系统架构

《智能系统学报》[ISSN1673-4785/CN23-1538/TP] 卷: 第15卷 期数: 2020年3期 页码: 468-474 栏目: 学术论文—智能系统 出版日期: 2020-05-05
Title:
System architecture of Texas Hold’em based on experience
作者:
高强1, 徐心和2, 王昊3, 白国力3, 曹瑞珉3
1. 沈阳大学 辽宁省装备制造综合自动化重点实验室，辽宁 沈阳 110044;
2. 东北大学 信息科学与工程学院，辽宁 沈阳 110819;
3. 东北大学 机械工程与自动化学院，辽宁 沈阳 110819
Author(s):
GAO Qiang1, XU Xinhe2, WANG Hao3, BAI Guoli3, CAO Ruimin3
1. Key Laboratory of Manufacturing Industrial Integrated Automation, Shenyang University, Shenyang 110044, China;
2. College of Information Science and Engineering, Northeastern University, Shenyang 110819, China;
3. School of Mechanical Engineering and Automation, Northeastern University, Shenyang 110819, China
关键词:
二人赌注无上限德州扑克; 计算机博弈; 非完全信息动态博弈; 博弈树; 深度学习; 专家库; 哈希表; 博弈策略
Keywords:
Heads-up no-limit Texas Hold’em; computer games; dynamic game with imperfect information; game tree; deep learning; expert database; Hash table; game strategy
分类号:
TP301.5
DOI:
10.11992/tis.201803043
摘要:
为了利用历史经验知识提高德州扑克博弈水平，提出一种二人赌注无上限的德州扑克博弈系统架构：对于知识库模块，利用海量历史牌局训练得到基于CNN的深度学习网络模型并构建了一个专家经验库；在系统的搜索模块中，构建了一种分阶段的德州扑克博弈树，利用专家经验和历史经验引导德州扑克博弈树的展开；对于系统的估值核心模块，构建了一种基于哈希技术的牌型对照表，以提高系统判定胜负的效率。实验结果表明本文提出的博弈系统架构具有更高的对弈水平。
Abstract:
To improve the level of Texas Hold’em through historical experience, this paper proposes a system architecture of heads-up no-limit Texas Hold’em for the knowledge base module. Mass historic games are used to train the deep learning network based on convolutional neural network, and an expert database is constructed for the search module of the system. Texas Hold’em structured game tree is developed and extended, and it is applied in terms of the expertise and historical experience to the core module for evaluation. A hand-ranking hash-based table is built to reduce the time required to evaluate hand rankings. The experimental result shows a higher playing level for the proposed system architecture.
参考文献/References:
[1] OSBORNE M J, RUBINSTEIN A. A course in game theory[M]. Cambridge: MIT Press, 1994.
[2] 胡裕靖, 高阳. 扑克游戏中的不完美信息博弈[J]. 中国计算机学会通讯, 2014, 10(9): 37-42
HU Yujing, GAO Yang. Games with incomplete information in Pokers[J]. China computer society newsletter, 2014, 10(9): 37-42
[3] LITTMAN M, ZINKEVICH M. The 2006 AAAI computer-poker competition[J]. ICGA journal, 2006, 29(3): 166-167.
[4] HARRIS M. The first “Man-Machine Poker Championship” begins tomorrow[N]. Poker News, 2007-07-22.
[5] BOWLING M, BURCH N, JOHANSON M, et al. Heads-up limit hold’em poker is solved[J]. Science, 2015, 347(6218): 145-149.
[6] BLAIR J R S, MUTCHLER D, LIU C. Games with imperfect information[R]. AAAI Technical Report FS-93-02. American: AAAI, 1993.
[7] HINTZE H. Libratus scores convincing sweep in man v. machine poker match[N]. Misc, News, 2017-01-31.
[8] BILLINGS D, DAVIDSON A, SCHAEFFER J, et al. The challenge of poker[J]. Artificial intelligence, 2002, 134(1/2): 201-240.
[9] BILLINGS D. Algorithms and assessment in computer poker[D]. Alberta: University of Alberta, 2006.
[10] ZOBRIST A L. A new hashing method with application for game playing[R]. Madison, USA: University of Wisconsin, 1970.
[11] WANG Jiao, LI Sizhong, XU Xinhe. A minors hash table in Chinese-chess programs2[J]. ICGA journal, 2010, 33(1): 18-33.
[12] BREUKER D M, UITERWIJK J W H M, VAN DEN HERIK H J. Replacement schemes for transposition tables[J]. ICGA journal, 1994, 17(4): 183-193.
[13] BREUKER D M, UITERWIJK J W H M, HERIK H J. Information in transposition tables[J]. Advances in computer chess, 1997, 27: 199-211.
[14] NELSON B L. Hash tables in Cray blitz[J]. ICGA journal, 1985, 8(1): 3-13.
[15] HYATT R M, COZZIE A. The effect of hash signature collisions in a chess program[J]. ICGA journal, 2005, 28(3): 131-139.
[16] TENENBAUM A M, LANGSAM Y, AUGENSTEIN M J. Data structures using C[M]. Englewood Cliffs: Prentice Hall, 1990: 456-461, 472.
[17] DENG Li, YU Dong. Deep learning: methods and applications[J]. Foundations and trends in signal processing, 2014, 7(3): 197-387.
[18] BENGIO Y. Learning deep architectures for AI[M]. Hanover: Now Publishers Inc., 2009.
[19] FUKUSHIMA K. Neocognitron: a self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position[J]. Biological cybernetics, 1980, 36(4): 193-202.
[20] YAKOVENKO N, CAO Liangliang, RAFFEL C, et al. Poker-CNN: a pattern learning strategy for making draws and bets in poker games[C]. Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence. Phoenix, USA, 2016: 360-367.
相似文献/References:
[1]魏钦刚,王 骄,徐心和,等.中国象棋计算机博弈开局库研究与设计[J].智能系统学报,2007,2(1):85.
　WEI Qin gang,WANG Jiao,XU Xin he,et al.A study and design of openingbook of computer Chinese C h ess[J].CAAI Transactions on Intelligent Systems,2007,2(3):85.
[2]黄 晨.棋类游戏中的先行权[J].智能系统学报,2007,2(3):91.
　HUANG Chen.The firstmove advantage in board games[J].CAAI Transactions on Intelligent Systems,2007,2(3):91.
[3]王亚杰,邱虹坤,吴燕燕,等.计算机博弈的研究与发展[J].智能系统学报,2016,11(6):788.[doi:10.11992/tis.201609006]
　WANG Yajie,QIU Hongkun,WU Yanyan,et al.Research and development of computer games[J].CAAI Transactions on Intelligent Systems,2016,11(3):788.[doi:10.11992/tis.201609006]
[4]李霞丽,吴立成,李永集.基于棋型的藏族“久”棋计算机博弈研究[J].智能系统学报,2018,13(4):577.[doi:10.11992/tis.201609023]
　LI Xiali,WU Licheng,LI Yongji.Tibetan JIU computer game research based on chess form[J].CAAI Transactions on Intelligent Systems,2018,13(3):577.[doi:10.11992/tis.201609023]
[5]魏印福,李舟军.动态规划求解中国象棋状态总数[J].智能系统学报,2019,14(1):108.[doi:10.11992/tis.201803008]
　WEI Yinfu,LI Zhoujun.A method for calculating the total number of states of Chinese chess on the basis of dynamic programming[J].CAAI Transactions on Intelligent Systems,2019,14(3):108.[doi:10.11992/tis.201803008]
[6]李淑琴,陈子鹏,郑蓝舟,等.竞技二打一游戏中同等牌力的研究[J].智能系统学报,2021,16(3):466.[doi:10.11992/tis.202007005]
　LI Shuqin,CHEN Zipeng,ZHENG Lanzhou,et al.Research on the equal card force competition system of competitive two against one game[J].CAAI Transactions on Intelligent Systems,2021,16(3):466.[doi:10.11992/tis.202007005]

备注/Memo

收稿日期:2018-03-26。
基金项目:辽宁省自然科学基金项目(20180550146，20170520386)
作者简介:高强，讲师，博士，主要研究方向为机器博弈、计算复杂性理论;徐心和，教授，博士生导师，中国人工智能学会常务理事，主要研究方向为控制理论与应用、系统仿真、智能机器人、机器博弈。主持完成国家自然科学基金、863基金、国家“八五”、“九五”攻关课题13项，其中8项通过省、部级鉴定，获科技进步奖国家三等1项，省部级科技进步奖多项。发表学术论文300余篇;王昊，博士研究生，主要研究方向为机器博弈
通讯作者:高强.E-mail:tommy_06@163.com

更新日期/Last Update: 1900-01-01
Copyright @ 《 智能系统学报》 编辑部
地址：（150001）黑龙江省哈尔滨市南岗区南通大街145-1号楼 电话：0451- 82534001、82518134