基于强化学习的桥牌叫牌策略研究
首发时间:2019-11-22
摘要:随着人工智能理论的发展,人类在越来越多的游戏领域被人工智能打败。而定约桥牌作为棋牌类游戏中规则最为复杂的游戏,对于目前的人工智能来说仍然是难以攻克的课题,因此,研究人工智能与定约桥牌的结合是十分有意义的。本文针对定约桥牌的叫牌阶段,利用强化学习方法对叫牌过程进行研究。通过大量机器人对打的桥牌数据来训练得到的基本模型,利用PolicyGradient等相关强化学习方法对基本模型进行增强,同时引入一种叫牌过滤机制来加快强化学习模型的收敛,并对训练结果和最终的叫牌能力进行分析。
For information in English, please click here
Bridge Bidbing based on Reinforcement Learning
Abstract:With the development of artificial intelligence theory, human beings have been defeated by artificial intelligence in more and more game fields. As the most complicated game in the chess game, the contract bridge is still a difficult task for the current artificial intelligence. Therefore, it is very meaningful to study the combination of artificial intelligence and contract bridge. In this paper, the intensive learning method is used to study the bidding process for the bidding stage of the contract bridge. The basic model is trained by a large number of robots playing bridge data, and the basic model is enhanced by using the related reinforcement learning method such as Policy Gradient. At the same time, a bid filtering mechanism is introduced to accelerate the convergence of the reinforcement learning model, and the training results and Final bidding ability analysis
Keywords: artificial intelligence contract bridge reinforcement learning
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
基于强化学习的桥牌叫牌策略研究
评论
全部评论0/1000