Comparing Strategies to Play a 2-Sided Dominoes Game

Andre R. Da Cruz, F. Guimarães, R. Takahashi
{"title":"Comparing Strategies to Play a 2-Sided Dominoes Game","authors":"Andre R. Da Cruz, F. Guimarães, R. Takahashi","doi":"10.1109/BRICS-CCI-CBIC.2013.59","DOIUrl":null,"url":null,"abstract":"This work presents four agents with different strategies to play a version of the 2-sided dominoes game, usually played in Minas Gerais state, Brazil. This incomplete information game must be played with two players and the goal is to discard all tiles first according to the rules. Each pair of agents was tested in a computational experiment, for 1,000,000 matches, in order to evaluate the individual effectiveness. In the first strategy, the agent uses random rules to select an adequate tile, the second agent observes the tiles already on the table and on its hand and selects one using a simple probability information computed in an amateur way, the third strategy also observes the tiles on the table and on the hand, and computes a probability information using the two end tiles of the table and the candidates opposite values in order to decide which one must be thrown, in the last strategy, the agent uses the third strategy and the Boltzmann exploration with a roulette wheel to select the tile. The results showed that the last strategy is the best and that even the random strategy is capable to win a significant number of matches.","PeriodicalId":306195,"journal":{"name":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2013-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 BRICS Congress on Computational Intelligence and 11th Brazilian Congress on Computational Intelligence","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/BRICS-CCI-CBIC.2013.59","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

Abstract

This work presents four agents with different strategies to play a version of the 2-sided dominoes game, usually played in Minas Gerais state, Brazil. This incomplete information game must be played with two players and the goal is to discard all tiles first according to the rules. Each pair of agents was tested in a computational experiment, for 1,000,000 matches, in order to evaluate the individual effectiveness. In the first strategy, the agent uses random rules to select an adequate tile, the second agent observes the tiles already on the table and on its hand and selects one using a simple probability information computed in an amateur way, the third strategy also observes the tiles on the table and on the hand, and computes a probability information using the two end tiles of the table and the candidates opposite values in order to decide which one must be thrown, in the last strategy, the agent uses the third strategy and the Boltzmann exploration with a roulette wheel to select the tile. The results showed that the last strategy is the best and that even the random strategy is capable to win a significant number of matches.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
比较双方多米诺骨牌游戏的策略
这项工作提出了四个具有不同策略的代理来玩通常在巴西米纳斯吉拉斯州进行的2面多米诺骨牌游戏的一个版本。这种不完全信息游戏必须由两名玩家进行,根据规则,目标是先丢弃所有的瓷砖。每对代理都在计算实验中进行了测试,进行了1,000,000次匹配,以评估个体有效性。在第一个策略,代理使用随机规则来选择一个适当的瓷砖,第二剂观察其手上的瓷砖已经放在桌子上,并选择一个使用一个简单的概率信息以业余方式计算,第三个战略还观察到瓷砖放在桌子上的手,和计算概率信息使用桌子的两端瓷砖和相对价值来决定哪一个候选人必须抛出,在过去的策略,代理使用第三种策略和波尔兹曼探索与轮盘赌轮选择瓷砖。结果表明,最后一种策略是最好的,即使是随机策略也能在相当数量的比赛中获胜。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Computer Simulations of Small Societies Under Social Transfer Systems Modeling of Reasoning in Intelligent Systems by Means of Integration of Methods Based on Case-Based Reasoning and Inductive Notions Formation Bi-dimensional Neural Equalizer Applied to Optical Receiver A New Algorithm Based on Differential Evolution for Combinatorial Optimization A Cooperative Parallel Particle Swarm Optimization for High-Dimension Problems on GPUs
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1