A MULTI-AGENT LOCAL-LEARNING ALGORITHM UNDER GROUP ENVIROMENT

A MULTI-AGENT LOCAL-LEARNING ALGORITHM UNDER GROUP ENVIROMENT

In this paper,a local-learning algorithm for multi-agent is presented based on the fact that individual agent performs local perception and local interaction under group environment.As for in-dividual-learning,agent adopts greedy strategy to maximize its reward when interacting with envi-ronment.In...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of electronics (China) 2009, Vol.26 (2), p.229-236
Hauptverfasser: Jiang, Daoping, Yin, Yixin, Ban, Xiaojuan, Meng, Xiangsong
Format: Artikel
Sprache:eng
Schlagworte:
Electrical Engineering
Engineering
Q学习算法
代理人
多Agent
相互作用
计算复杂度
贪婪策略
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 236
container_issue 2
container_start_page 229
container_title Journal of electronics (China)
container_volume 26
creator Jiang, Daoping
Yin, Yixin
Ban, Xiaojuan
Meng, Xiangsong
description In this paper,a local-learning algorithm for multi-agent is presented based on the fact that individual agent performs local perception and local interaction under group environment.As for in-dividual-learning,agent adopts greedy strategy to maximize its reward when interacting with envi-ronment.In group-learning,local interaction takes place between each two agents.A local-learning algorithm to choose and modify agents＇ actions is proposed to improve the traditional Q-learning algorithm,respectively in the situations of zero-sum games and general-sum games with unique equi-librium or multi-equilibrium.And this local-learning algorithm is proved to be convergent and the computation complexity is lower than the Nash-Q.Additionally,through grid-game test,it is indicated that by using this local-learning algorithm,the local behaviors of agents can spread to globe.
doi_str_mv 10.1007/s11767-007-0163-4
format Article
fullrecord <record><control><sourceid>wanfang_jour_cross</sourceid><recordid>TN_cdi_wanfang_journals_dzkxxk_e200902012</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><cqvip_id>29986849</cqvip_id><wanfj_id>dzkxxk_e200902012</wanfj_id><sourcerecordid>dzkxxk_e200902012</sourcerecordid><originalsourceid>FETCH-LOGICAL-c2162-e70d9f9989bc371e32e7c66188291ebac2ab9c9eeab2c3ee38ed491945f38e663</originalsourceid><addsrcrecordid>eNp9kE9Pg0AUxDdGE2v1A3gj3jysvrcLC3vwQCpSEgqGgNcN0KX2j1QhxuqndwlNvHl6c5jfTN4Qco1whwDufY_oCpcaSQEFp_YJmaCUnIJA55RMgKFLpcfYObno-w2Awz0HJuTBtxZFnEfUD4Mkt-J05sc0DvwsiZLQ8uMwzaJ8vrCK5DHIrDBLi2crSF6iLF0Y_yU5a8pdr6-Od0qKpyCfzWmchtGQVDMUjGoXlrKR0pNVzV3UnGm3FgI9j0nUVVmzspK11LqsWM215p5e2hKl7TRGCsGn5HbM_SrbpmxXarP_7FrTqJY_28NhqzQDkMAAmfHi6K27fd93ulHv3fqt7L4Vghq2UuNWapDDVso2DBuZ3njble7-Cv6Dbo5Fr_t29WE4ZV7ZNuudVsw8Kzxb8l_aY3Ex</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>A MULTI-AGENT LOCAL-LEARNING ALGORITHM UNDER GROUP ENVIROMENT</title><source>Alma/SFX Local Collection</source><creator>Jiang, Daoping ; Yin, Yixin ; Ban, Xiaojuan ; Meng, Xiangsong</creator><creatorcontrib>Jiang, Daoping ; Yin, Yixin ; Ban, Xiaojuan ; Meng, Xiangsong</creatorcontrib><description>In this paper,a local-learning algorithm for multi-agent is presented based on the fact that individual agent performs local perception and local interaction under group environment.As for in-dividual-learning,agent adopts greedy strategy to maximize its reward when interacting with envi-ronment.In group-learning,local interaction takes place between each two agents.A local-learning algorithm to choose and modify agents＇ actions is proposed to improve the traditional Q-learning algorithm,respectively in the situations of zero-sum games and general-sum games with unique equi-librium or multi-equilibrium.And this local-learning algorithm is proved to be convergent and the computation complexity is lower than the Nash-Q.Additionally,through grid-game test,it is indicated that by using this local-learning algorithm,the local behaviors of agents can spread to globe.</description><identifier>ISSN: 0217-9822</identifier><identifier>EISSN: 1993-0615</identifier><identifier>DOI: 10.1007/s11767-007-0163-4</identifier><language>eng</language><publisher>Heidelberg: SP Science Press</publisher><subject>Electrical Engineering ; Engineering ; Q学习算法 ; 代理人 ; 多Agent ; 相互作用 ; 计算复杂度 ; 贪婪策略</subject><ispartof>Journal of electronics (China), 2009, Vol.26 (2), p.229-236</ispartof><rights>Science Press, Institute of Electronics, CAS and Springer-Verlag GmbH 2009</rights><rights>Copyright © Wanfang Data Co. Ltd. All Rights Reserved.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c2162-e70d9f9989bc371e32e7c66188291ebac2ab9c9eeab2c3ee38ed491945f38e663</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Uhttp://image.cqvip.com/vip1000/qk/85266X/85266X.jpg</thumbnail><link.rule.ids>314,776,780,27903,27904</link.rule.ids></links><search><creatorcontrib>Jiang, Daoping</creatorcontrib><creatorcontrib>Yin, Yixin</creatorcontrib><creatorcontrib>Ban, Xiaojuan</creatorcontrib><creatorcontrib>Meng, Xiangsong</creatorcontrib><title>A MULTI-AGENT LOCAL-LEARNING ALGORITHM UNDER GROUP ENVIROMENT</title><title>Journal of electronics (China)</title><addtitle>J. Electron.(China)</addtitle><addtitle>Journal of Electronics</addtitle><description>In this paper,a local-learning algorithm for multi-agent is presented based on the fact that individual agent performs local perception and local interaction under group environment.As for in-dividual-learning,agent adopts greedy strategy to maximize its reward when interacting with envi-ronment.In group-learning,local interaction takes place between each two agents.A local-learning algorithm to choose and modify agents＇ actions is proposed to improve the traditional Q-learning algorithm,respectively in the situations of zero-sum games and general-sum games with unique equi-librium or multi-equilibrium.And this local-learning algorithm is proved to be convergent and the computation complexity is lower than the Nash-Q.Additionally,through grid-game test,it is indicated that by using this local-learning algorithm,the local behaviors of agents can spread to globe.</description><subject>Electrical Engineering</subject><subject>Engineering</subject><subject>Q学习算法</subject><subject>代理人</subject><subject>多Agent</subject><subject>相互作用</subject><subject>计算复杂度</subject><subject>贪婪策略</subject><issn>0217-9822</issn><issn>1993-0615</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2009</creationdate><recordtype>article</recordtype><recordid>eNp9kE9Pg0AUxDdGE2v1A3gj3jysvrcLC3vwQCpSEgqGgNcN0KX2j1QhxuqndwlNvHl6c5jfTN4Qco1whwDufY_oCpcaSQEFp_YJmaCUnIJA55RMgKFLpcfYObno-w2Awz0HJuTBtxZFnEfUD4Mkt-J05sc0DvwsiZLQ8uMwzaJ8vrCK5DHIrDBLi2crSF6iLF0Y_yU5a8pdr6-Od0qKpyCfzWmchtGQVDMUjGoXlrKR0pNVzV3UnGm3FgI9j0nUVVmzspK11LqsWM215p5e2hKl7TRGCsGn5HbM_SrbpmxXarP_7FrTqJY_28NhqzQDkMAAmfHi6K27fd93ulHv3fqt7L4Vghq2UuNWapDDVso2DBuZ3njble7-Cv6Dbo5Fr_t29WE4ZV7ZNuudVsw8Kzxb8l_aY3Ex</recordid><startdate>2009</startdate><enddate>2009</enddate><creator>Jiang, Daoping</creator><creator>Yin, Yixin</creator><creator>Ban, Xiaojuan</creator><creator>Meng, Xiangsong</creator><general>SP Science Press</general><general>School of Information Engineering, University of Science &amp; Technology Beijing, Beijing 100083, China</general><scope>2RA</scope><scope>92L</scope><scope>CQIGP</scope><scope>W92</scope><scope>~WA</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>2B.</scope><scope>4A8</scope><scope>92I</scope><scope>93N</scope><scope>PSX</scope><scope>TCJ</scope></search><sort><creationdate>2009</creationdate><title>A MULTI-AGENT LOCAL-LEARNING ALGORITHM UNDER GROUP ENVIROMENT</title><author>Jiang, Daoping ; Yin, Yixin ; Ban, Xiaojuan ; Meng, Xiangsong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c2162-e70d9f9989bc371e32e7c66188291ebac2ab9c9eeab2c3ee38ed491945f38e663</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2009</creationdate><topic>Electrical Engineering</topic><topic>Engineering</topic><topic>Q学习算法</topic><topic>代理人</topic><topic>多Agent</topic><topic>相互作用</topic><topic>计算复杂度</topic><topic>贪婪策略</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Jiang, Daoping</creatorcontrib><creatorcontrib>Yin, Yixin</creatorcontrib><creatorcontrib>Ban, Xiaojuan</creatorcontrib><creatorcontrib>Meng, Xiangsong</creatorcontrib><collection>中文科技期刊数据库</collection><collection>中文科技期刊数据库-CALIS站点</collection><collection>中文科技期刊数据库-7.0平台</collection><collection>中文科技期刊数据库-工程技术</collection><collection>中文科技期刊数据库- 镜像站点</collection><collection>CrossRef</collection><collection>Wanfang Data Journals - Hong Kong</collection><collection>WANFANG Data Centre</collection><collection>Wanfang Data Journals</collection><collection>万方数据期刊 - 香港版</collection><collection>China Online Journals (COJ)</collection><collection>China Online Journals (COJ)</collection><jtitle>Journal of electronics (China)</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Jiang, Daoping</au><au>Yin, Yixin</au><au>Ban, Xiaojuan</au><au>Meng, Xiangsong</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>A MULTI-AGENT LOCAL-LEARNING ALGORITHM UNDER GROUP ENVIROMENT</atitle><jtitle>Journal of electronics (China)</jtitle><stitle>J. Electron.(China)</stitle><addtitle>Journal of Electronics</addtitle><date>2009</date><risdate>2009</risdate><volume>26</volume><issue>2</issue><spage>229</spage><epage>236</epage><pages>229-236</pages><issn>0217-9822</issn><eissn>1993-0615</eissn><abstract>In this paper,a local-learning algorithm for multi-agent is presented based on the fact that individual agent performs local perception and local interaction under group environment.As for in-dividual-learning,agent adopts greedy strategy to maximize its reward when interacting with envi-ronment.In group-learning,local interaction takes place between each two agents.A local-learning algorithm to choose and modify agents＇ actions is proposed to improve the traditional Q-learning algorithm,respectively in the situations of zero-sum games and general-sum games with unique equi-librium or multi-equilibrium.And this local-learning algorithm is proved to be convergent and the computation complexity is lower than the Nash-Q.Additionally,through grid-game test,it is indicated that by using this local-learning algorithm,the local behaviors of agents can spread to globe.</abstract><cop>Heidelberg</cop><pub>SP Science Press</pub><doi>10.1007/s11767-007-0163-4</doi><tpages>8</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0217-9822
ispartof Journal of electronics (China), 2009, Vol.26 (2), p.229-236
issn 0217-9822
1993-0615
language eng
recordid cdi_wanfang_journals_dzkxxk_e200902012
source Alma/SFX Local Collection
subjects Electrical Engineering
Engineering
Q学习算法
代理人
多Agent
相互作用
计算复杂度
贪婪策略
title A MULTI-AGENT LOCAL-LEARNING ALGORITHM UNDER GROUP ENVIROMENT
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-22T21%3A02%3A39IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-wanfang_jour_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=A%20MULTI-AGENT%20LOCAL-LEARNING%20ALGORITHM%20UNDER%20GROUP%20ENVIROMENT&rft.jtitle=Journal%20of%20electronics%20(China)&rft.au=Jiang,%20Daoping&rft.date=2009&rft.volume=26&rft.issue=2&rft.spage=229&rft.epage=236&rft.pages=229-236&rft.issn=0217-9822&rft.eissn=1993-0615&rft_id=info:doi/10.1007/s11767-007-0163-4&rft_dat=%3Cwanfang_jour_cross%3Edzkxxk_e200902012%3C/wanfang_jour_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_cqvip_id=29986849&rft_wanfj_id=dzkxxk_e200902012&rfr_iscdi=true