基于隐式对手建模的多智能体学习

Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600) Pub Date : 2002-05-12 DOI:10.1109/CEC.2002.1004470

Ronald V. Bjarnason, T. Peterson

{"title":"基于隐式对手建模的多智能体学习","authors":"Ronald V. Bjarnason, T. Peterson","doi":"10.1109/CEC.2002.1004470","DOIUrl":null,"url":null,"abstract":"We present a learning algorithm for two player stochastic games. The algorithm generates optimal deterministic finite automata (DFA) strategies against opponents who can be modeled by probabilistic action automata. The algorithm generates dynamic history trees based on statistical tests to eliminate state aliasing. Experiments are conducted in an iterated prisoner's dilemma environment.","PeriodicalId":184547,"journal":{"name":"Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2002-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":"{\"title\":\"Multi-agent learning via implicit opponent modeling\",\"authors\":\"Ronald V. Bjarnason, T. Peterson\",\"doi\":\"10.1109/CEC.2002.1004470\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We present a learning algorithm for two player stochastic games. The algorithm generates optimal deterministic finite automata (DFA) strategies against opponents who can be modeled by probabilistic action automata. The algorithm generates dynamic history trees based on statistical tests to eliminate state aliasing. Experiments are conducted in an iterated prisoner's dilemma environment.\",\"PeriodicalId\":184547,\"journal\":{\"name\":\"Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600)\",\"volume\":\"14 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2002-05-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"7\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CEC.2002.1004470\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CEC.2002.1004470","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 7

摘要

提出了一种二人随机博弈的学习算法。该算法生成针对对手的最优确定性有限自动机(DFA)策略，这些策略可以用概率动作自动机建模。该算法基于统计测试生成动态历史树，消除状态混叠。实验在反复的囚徒困境环境中进行。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Multi-agent learning via implicit opponent modeling

We present a learning algorithm for two player stochastic games. The algorithm generates optimal deterministic finite automata (DFA) strategies against opponents who can be modeled by probabilistic action automata. The algorithm generates dynamic history trees based on statistical tests to eliminate state aliasing. Experiments are conducted in an iterated prisoner's dilemma environment.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600)

自引率

0.00%

发文量