Social Agents Playing a Periodical Policy
Identifieur interne : 001674 ( Istex/Corpus ); précédent : 001673; suivant : 001675Social Agents Playing a Periodical Policy
Auteurs : Ann Nowé ; Johan Parent ; Katja VerbeeckSource :
- Lecture Notes in Computer Science [ 0302-9743 ] ; 2001.
Abstract
Abstract: Coordination is an important issue in multiagent systems. Within the stochastic game framework this problem translates to policy learning in a joint action space. This technique however suffers some important drawbacks like the assumption of the existence of a unique Nash equilibrium and synchronicity, the need for central control, the cost of communication, etc. Moreover in general sum games it is not always clear which policies should be learned. Playing pure Nash equilibrium is often unfair to at least one of the players, while playing a mixed strategy doesn’t give any guarantee for coordination and usually results in a sub-optimal payoff for all agents. In this work we show the usefulness of periodical policies, which arise as a side effect of the fairness conditions used by the agents. We are interested in games which assume competition between the players, but where the overall performance can only be as good as the performance of the poorest player. Players are social distributed reinforcement learners, who have to learn to equalize their payoff. Our approach is illustrated on synchronous one-step games as well as on asynchronous job scheduling games.
Url:
DOI: 10.1007/3-540-44795-4_33
Links to Exploration step
ISTEX:B9569E9E27EC810019E7B082F8F573E317E9C04ALe document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct:series"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Social Agents Playing a Periodical Policy</title><author><name sortKey="Nowe, Ann" sort="Nowe, Ann" uniqKey="Nowe A" first="Ann" last="Nowé">Ann Nowé</name><affiliation><mods:affiliation>Computational Modeling Lab, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>Vrije Universiteit Brussel, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: asnowe@info.vub.ac.be</mods:affiliation></affiliation></author><author><name sortKey="Parent, Johan" sort="Parent, Johan" uniqKey="Parent J" first="Johan" last="Parent">Johan Parent</name><affiliation><mods:affiliation>Computational Modeling Lab, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>Vrije Universiteit Brussel, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: johan@info.vub.ac.be</mods:affiliation></affiliation></author><author><name sortKey="Verbeeck, Katja" sort="Verbeeck, Katja" uniqKey="Verbeeck K" first="Katja" last="Verbeeck">Katja Verbeeck</name><affiliation><mods:affiliation>Computational Modeling Lab, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>Vrije Universiteit Brussel, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: kaverbee@vub.ac.be</mods:affiliation></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:B9569E9E27EC810019E7B082F8F573E317E9C04A</idno><date when="2001" year="2001">2001</date><idno type="doi">10.1007/3-540-44795-4_33</idno><idno type="url">https://api.istex.fr/document/B9569E9E27EC810019E7B082F8F573E317E9C04A/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">001674</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Social Agents Playing a Periodical Policy</title><author><name sortKey="Nowe, Ann" sort="Nowe, Ann" uniqKey="Nowe A" first="Ann" last="Nowé">Ann Nowé</name><affiliation><mods:affiliation>Computational Modeling Lab, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>Vrije Universiteit Brussel, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: asnowe@info.vub.ac.be</mods:affiliation></affiliation></author><author><name sortKey="Parent, Johan" sort="Parent, Johan" uniqKey="Parent J" first="Johan" last="Parent">Johan Parent</name><affiliation><mods:affiliation>Computational Modeling Lab, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>Vrije Universiteit Brussel, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: johan@info.vub.ac.be</mods:affiliation></affiliation></author><author><name sortKey="Verbeeck, Katja" sort="Verbeeck, Katja" uniqKey="Verbeeck K" first="Katja" last="Verbeeck">Katja Verbeeck</name><affiliation><mods:affiliation>Computational Modeling Lab, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>Vrije Universiteit Brussel, Belgium</mods:affiliation></affiliation><affiliation><mods:affiliation>E-mail: kaverbee@vub.ac.be</mods:affiliation></affiliation></author></analytic><monogr></monogr><series><title level="s">Lecture Notes in Computer Science</title><imprint><date>2001</date></imprint><idno type="ISSN">0302-9743</idno></series><idno type="istex">B9569E9E27EC810019E7B082F8F573E317E9C04A</idno><idno type="DOI">10.1007/3-540-44795-4_33</idno><idno type="ChapterID">Chap33</idno><idno type="ChapterID">33</idno></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">Abstract: Coordination is an important issue in multiagent systems. Within the stochastic game framework this problem translates to policy learning in a joint action space. This technique however suffers some important drawbacks like the assumption of the existence of a unique Nash equilibrium and synchronicity, the need for central control, the cost of communication, etc. Moreover in general sum games it is not always clear which policies should be learned. Playing pure Nash equilibrium is often unfair to at least one of the players, while playing a mixed strategy doesn’t give any guarantee for coordination and usually results in a sub-optimal payoff for all agents. In this work we show the usefulness of periodical policies, which arise as a side effect of the fairness conditions used by the agents. We are interested in games which assume competition between the players, but where the overall performance can only be as good as the performance of the poorest player. Players are social distributed reinforcement learners, who have to learn to equalize their payoff. Our approach is illustrated on synchronous one-step games as well as on asynchronous job scheduling games.</div></front></TEI><istex><corpusName>springer</corpusName><author><json:item><name>Ann Nowé</name><affiliations><json:string>Computational Modeling Lab, Belgium</json:string><json:string>Vrije Universiteit Brussel, Belgium</json:string><json:string>E-mail: asnowe@info.vub.ac.be</json:string></affiliations></json:item><json:item><name>Johan Parent</name><affiliations><json:string>Computational Modeling Lab, Belgium</json:string><json:string>Vrije Universiteit Brussel, Belgium</json:string><json:string>E-mail: johan@info.vub.ac.be</json:string></affiliations></json:item><json:item><name>Katja Verbeeck</name><affiliations><json:string>Computational Modeling Lab, Belgium</json:string><json:string>Vrije Universiteit Brussel, Belgium</json:string><json:string>E-mail: kaverbee@vub.ac.be</json:string></affiliations></json:item></author><language><json:string>eng</json:string></language><abstract>Abstract: Coordination is an important issue in multiagent systems. Within the stochastic game framework this problem translates to policy learning in a joint action space. This technique however suffers some important drawbacks like the assumption of the existence of a unique Nash equilibrium and synchronicity, the need for central control, the cost of communication, etc. Moreover in general sum games it is not always clear which policies should be learned. Playing pure Nash equilibrium is often unfair to at least one of the players, while playing a mixed strategy doesn’t give any guarantee for coordination and usually results in a sub-optimal payoff for all agents. In this work we show the usefulness of periodical policies, which arise as a side effect of the fairness conditions used by the agents. We are interested in games which assume competition between the players, but where the overall performance can only be as good as the performance of the poorest player. Players are social distributed reinforcement learners, who have to learn to equalize their payoff. Our approach is illustrated on synchronous one-step games as well as on asynchronous job scheduling games.</abstract><qualityIndicators><score>6.215</score><pdfVersion>1.3</pdfVersion><pdfPageSize>430 x 650 pts</pdfPageSize><refBibsNative>false</refBibsNative><keywordCount>0</keywordCount><abstractCharCount>1186</abstractCharCount><pdfWordCount>3971</pdfWordCount><pdfCharCount>21711</pdfCharCount><pdfPageCount>12</pdfPageCount><abstractWordCount>187</abstractWordCount></qualityIndicators><title>Social Agents Playing a Periodical Policy</title><genre><json:string>Book Chapter</json:string></genre><serie><editor><json:item><name>G. Goos</name></json:item><json:item><name>J. Hartmanis</name></json:item><json:item><name>J. van Leeuwen</name></json:item></editor><issn><json:string>0302-9743</json:string></issn><genre><json:string>book series</json:string></genre><language><json:string>unknown</json:string></language><title>Lecture Notes in Computer Science</title><copyrightDate>2001</copyrightDate></serie><host><volume>2167</volume><editor><json:item><name>Luc De Raedt</name><affiliations><json:string>Department of Computer Science, Albert-Ludwigs University Freiburg, Georges Köhler-Allee, Geb. 079, 79110, Freiburg, Germany</json:string><json:string>E-mail: deraedt@informatik.uni-freiburg.de</json:string></affiliations></json:item><json:item><name>Peter Flach</name><affiliations><json:string>Department of Computer Science, University of Bristol, Merchant Ventures Bldg., Woodland Road, BS8 1UB, Bristol, UK</json:string><json:string>E-mail: peter.flach@bristol.ac.uk</json:string></affiliations></json:item></editor><pages><last>393</last><first>382</first></pages><subject><json:item><value>Computer Science</value></json:item><json:item><value>Computer Science</value></json:item><json:item><value>Artificial Intelligence (incl. Robotics)</value></json:item><json:item><value>Algorithm Analysis and Problem Complexity</value></json:item><json:item><value>Mathematical Logic and Formal Languages</value></json:item></subject><isbn><json:string>978-3-540-42536-6</json:string></isbn><genre><json:string>book series</json:string></genre><language><json:string>unknown</json:string></language><eisbn><json:string>978-3-540-44795-5</json:string></eisbn><title>Machine Learning: ECML 2001</title><copyrightDate>2001</copyrightDate><doi><json:string>10.1007/3-540-44795-4</json:string></doi></host><copyrightDate>2001</copyrightDate><doi><json:string>10.1007/3-540-44795-4_33</json:string></doi><id>B9569E9E27EC810019E7B082F8F573E317E9C04A</id><fulltext><json:item><original>true</original><mimetype>application/pdf</mimetype><extension>pdf</extension><uri>https://api.istex.fr/document/B9569E9E27EC810019E7B082F8F573E317E9C04A/fulltext/pdf</uri></json:item><json:item><original>false</original><mimetype>application/zip</mimetype><extension>zip</extension><uri>https://api.istex.fr/document/B9569E9E27EC810019E7B082F8F573E317E9C04A/fulltext/zip</uri></json:item><istex:fulltextTEI uri="https://api.istex.fr/document/B9569E9E27EC810019E7B082F8F573E317E9C04A/fulltext/tei"><teiHeader><fileDesc><titleStmt><title level="a" type="main" xml:lang="en">Social Agents Playing a Periodical Policy</title><respStmt xml:id="ISTEX-API" resp="Références bibliographiques récupérées via GROBID" name="ISTEX-API (INIST-CNRS)"></respStmt></titleStmt><publicationStmt><authority>ISTEX</authority><publisher>Springer Berlin Heidelberg</publisher><pubPlace>Berlin, Heidelberg</pubPlace><date>2001</date></publicationStmt><sourceDesc><biblStruct type="inbook"><analytic><title level="a" type="main" xml:lang="en">Social Agents Playing a Periodical Policy</title><author><persName><forename type="first">Ann</forename><surname>Nowé</surname></persName><email>asnowe@info.vub.ac.be</email><affiliation>Computational Modeling Lab, Belgium</affiliation><affiliation>Vrije Universiteit Brussel, Belgium</affiliation></author><author><persName><forename type="first">Johan</forename><surname>Parent</surname></persName><email>johan@info.vub.ac.be</email><affiliation>Computational Modeling Lab, Belgium</affiliation><affiliation>Vrije Universiteit Brussel, Belgium</affiliation></author><author><persName><forename type="first">Katja</forename><surname>Verbeeck</surname></persName><email>kaverbee@vub.ac.be</email><affiliation>Computational Modeling Lab, Belgium</affiliation><affiliation>Vrije Universiteit Brussel, Belgium</affiliation></author></analytic><monogr><title level="m">Machine Learning: ECML 2001</title><title level="m" type="sub">12th European Conference on Machine Learning Freiburg, Germany, September 5–7, 2001 Proceedings</title><idno type="pISBN">978-3-540-42536-6</idno><idno type="eISBN">978-3-540-44795-5</idno><idno type="DOI">10.1007/3-540-44795-4</idno><idno type="BookID">3-540-44795-4</idno><idno type="BookTitleID">69457</idno><idno type="BookSequenceNumber">2167</idno><idno type="BookVolumeNumber">2167</idno><idno type="BookChapterCount">55</idno><editor><persName><forename type="first">Luc</forename><surname>De Raedt</surname></persName><email>deraedt@informatik.uni-freiburg.de</email><affiliation>Department of Computer Science, Albert-Ludwigs University Freiburg, Georges Köhler-Allee, Geb. 079, 79110, Freiburg, Germany</affiliation></editor><editor><persName><forename type="first">Peter</forename><surname>Flach</surname></persName><email>peter.flach@bristol.ac.uk</email><affiliation>Department of Computer Science, University of Bristol, Merchant Ventures Bldg., Woodland Road, BS8 1UB, Bristol, UK</affiliation></editor><imprint><publisher>Springer Berlin Heidelberg</publisher><pubPlace>Berlin, Heidelberg</pubPlace><date>2001</date><biblScope unit="volume">2167</biblScope><biblScope unit="page" from="382">382</biblScope><biblScope unit="page" to="393">393</biblScope></imprint></monogr><series><title level="s">Lecture Notes in Computer Science</title><editor><persName><forename type="first">G.</forename><surname>Goos</surname></persName></editor><editor><persName><forename type="first">J.</forename><surname>Hartmanis</surname></persName></editor><editor><persName><forename type="first">J.</forename><surname>van Leeuwen</surname></persName></editor><biblScope><date>2001</date></biblScope><idno type="pISSN">0302-9743</idno><idno type="seriesId">558</idno></series><series><title level="s">Lecture Notes in Artificial Intelligence</title><title level="s" type="sub">Subseries of Lecture Notes in Computer Science</title><editor><persName><forename type="first">G.</forename><surname>Goos</surname></persName></editor><editor><persName><forename type="first">J.</forename><surname>Hartmanis</surname></persName></editor><editor><persName><forename type="first">J.</forename><surname>van Leeuwen</surname></persName></editor><editor><persName><forename type="first">Luc</forename><surname>De Raedt</surname></persName><email>deraedt@informatik.uni-freiburg.de</email><affiliation>Department of Computer Science, Albert-Ludwigs University Freiburg, Georges Köhler-Allee, Geb. 079, 79110, Freiburg, Germany</affiliation></editor><editor><persName><forename type="first">Peter</forename><surname>Flach</surname></persName><email>peter.flach@bristol.ac.uk</email><affiliation>Department of Computer Science, University of Bristol, Merchant Ventures Bldg., Woodland Road, BS8 1UB, Bristol, UK</affiliation></editor><editor><persName><forename type="first">Jaime</forename><forename type="first">G.</forename><surname>Carbonell</surname></persName><affiliation>Carnegie Mellon University, Pittsburgh, PA, USA</affiliation></editor><editor><persName><forename type="first">Jörg</forename><surname>Siekmann</surname></persName><affiliation>University of Saarland, Saarbrücken, Germany</affiliation></editor><biblScope type="seriesId">1244</biblScope></series><idno type="istex">B9569E9E27EC810019E7B082F8F573E317E9C04A</idno><idno type="DOI">10.1007/3-540-44795-4_33</idno><idno type="ChapterID">Chap33</idno><idno type="ChapterID">33</idno></biblStruct></sourceDesc></fileDesc><profileDesc><creation><date>2001</date></creation><langUsage><language ident="en">en</language></langUsage><abstract xml:lang="en"><p>Abstract: Coordination is an important issue in multiagent systems. Within the stochastic game framework this problem translates to policy learning in a joint action space. This technique however suffers some important drawbacks like the assumption of the existence of a unique Nash equilibrium and synchronicity, the need for central control, the cost of communication, etc. Moreover in general sum games it is not always clear which policies should be learned. Playing pure Nash equilibrium is often unfair to at least one of the players, while playing a mixed strategy doesn’t give any guarantee for coordination and usually results in a sub-optimal payoff for all agents. In this work we show the usefulness of periodical policies, which arise as a side effect of the fairness conditions used by the agents. We are interested in games which assume competition between the players, but where the overall performance can only be as good as the performance of the poorest player. Players are social distributed reinforcement learners, who have to learn to equalize their payoff. Our approach is illustrated on synchronous one-step games as well as on asynchronous job scheduling games.</p></abstract><textClass><keywords scheme="Book Subject Collection"><list><label>SUCO11645</label><item><term>Computer Science</term></item></list></keywords></textClass><textClass><keywords scheme="Book Subject Group"><list><label>I</label><label>I21017</label><label>I16021</label><label>I16048</label><item><term>Computer Science</term></item><item><term>Artificial Intelligence (incl. Robotics)</term></item><item><term>Algorithm Analysis and Problem Complexity</term></item><item><term>Mathematical Logic and Formal Languages</term></item></list></keywords></textClass></profileDesc><revisionDesc><change>undefined</change><change>[object Object]</change></revisionDesc></teiHeader></istex:fulltextTEI><json:item><original>false</original><mimetype>text/plain</mimetype><extension>txt</extension><uri>https://api.istex.fr/document/B9569E9E27EC810019E7B082F8F573E317E9C04A/fulltext/txt</uri></json:item></fulltext><metadata><istex:metadataXml wicri:clean="Springer, Publisher found" wicri:toSee="no header"><istex:xmlDeclaration>version="1.0" encoding="UTF-8"</istex:xmlDeclaration><istex:docType PUBLIC="-//Springer-Verlag//DTD A++ V2.4//EN" URI="http://devel.springer.de/A++/V2.4/DTD/A++V2.4.dtd" name="istex:docType"></istex:docType><istex:document><Publisher><PublisherInfo><PublisherName>Springer Berlin Heidelberg</PublisherName><PublisherLocation>Berlin, Heidelberg</PublisherLocation></PublisherInfo><Series><SeriesInfo SeriesType="Series" TocLevels="0"><SeriesID>558</SeriesID><SeriesPrintISSN>0302-9743</SeriesPrintISSN><SeriesTitle Language="En">Lecture Notes in Computer Science</SeriesTitle></SeriesInfo><SeriesHeader><EditorGroup><Editor><EditorName DisplayOrder="Western"><GivenName>G.</GivenName><FamilyName>Goos</FamilyName></EditorName></Editor><Editor><EditorName DisplayOrder="Western"><GivenName>J.</GivenName><FamilyName>Hartmanis</FamilyName></EditorName></Editor><Editor><EditorName DisplayOrder="Western"><GivenName>J.</GivenName><Particle>van</Particle><FamilyName>Leeuwen</FamilyName></EditorName></Editor></EditorGroup></SeriesHeader><Book Language="En"><BookInfo BookProductType="Proceedings" ContainsESM="No" Language="En" MediaType="eBook" NumberingStyle="Unnumbered" TocLevels="0"><BookID>3-540-44795-4</BookID><BookTitle>Machine Learning: ECML 2001</BookTitle><BookSubTitle>12th European Conference on Machine Learning Freiburg, Germany, September 5–7, 2001 Proceedings</BookSubTitle><BookVolumeNumber>2167</BookVolumeNumber><BookSequenceNumber>2167</BookSequenceNumber><BookDOI>10.1007/3-540-44795-4</BookDOI><BookTitleID>69457</BookTitleID><BookPrintISBN>978-3-540-42536-6</BookPrintISBN><BookElectronicISBN>978-3-540-44795-5</BookElectronicISBN><BookChapterCount>55</BookChapterCount><BookHistory><OnlineDate><Year>2001</Year><Month>8</Month><Day>30</Day></OnlineDate></BookHistory><BookCopyright><CopyrightHolderName>Springer-Verlag Berlin Heidelberg</CopyrightHolderName><CopyrightYear>2001</CopyrightYear></BookCopyright><BookSubjectGroup><BookSubject Code="I" Type="Primary">Computer Science</BookSubject><BookSubject Code="I21017" Priority="1" Type="Secondary">Artificial Intelligence (incl. Robotics)</BookSubject><BookSubject Code="I16021" Priority="2" Type="Secondary">Algorithm Analysis and Problem Complexity</BookSubject><BookSubject Code="I16048" Priority="3" Type="Secondary">Mathematical Logic and Formal Languages</BookSubject><SubjectCollection Code="SUCO11645">Computer Science</SubjectCollection></BookSubjectGroup><BookContext><SeriesID>558</SeriesID></BookContext></BookInfo><BookHeader><EditorGroup><Editor AffiliationIDS="Aff1"><EditorName DisplayOrder="Western"><GivenName>Luc</GivenName><Particle>De</Particle><FamilyName>Raedt</FamilyName></EditorName><Contact><Email>deraedt@informatik.uni-freiburg.de</Email></Contact></Editor><Editor AffiliationIDS="Aff2"><EditorName DisplayOrder="Western"><GivenName>Peter</GivenName><FamilyName>Flach</FamilyName></EditorName><Contact><Email>peter.flach@bristol.ac.uk</Email></Contact></Editor><Affiliation ID="Aff1"><OrgDivision>Department of Computer Science</OrgDivision><OrgName>Albert-Ludwigs University Freiburg</OrgName><OrgAddress><Street>Georges Köhler-Allee, Geb. 079</Street><Postcode>79110</Postcode><City>Freiburg</City><Country>Germany</Country></OrgAddress></Affiliation><Affiliation ID="Aff2"><OrgDivision>Department of Computer Science</OrgDivision><OrgName>University of Bristol</OrgName><OrgAddress><Street>Merchant Ventures Bldg., Woodland Road</Street><City>Bristol</City><Postcode>BS8 1UB</Postcode><Country>UK</Country></OrgAddress></Affiliation></EditorGroup></BookHeader><Part ID="Part1"><PartInfo TocLevels="0"><PartID>1</PartID><PartSequenceNumber>1</PartSequenceNumber><PartTitle>Regular Papers</PartTitle><PartChapterCount>50</PartChapterCount><PartContext><SeriesID>558</SeriesID><BookID>3-540-44795-4</BookID><BookTitle>Machine Learning: ECML 2001</BookTitle></PartContext></PartInfo><Chapter ID="Chap33" Language="En"><ChapterInfo ChapterType="OriginalPaper" ContainsESM="No" Language="En" NumberingStyle="Unnumbered" TocLevels="0"><ChapterID>33</ChapterID><ChapterDOI>10.1007/3-540-44795-4_33</ChapterDOI><ChapterSequenceNumber>33</ChapterSequenceNumber><ChapterTitle Language="En">Social Agents Playing a Periodical Policy</ChapterTitle><ChapterFirstPage>382</ChapterFirstPage><ChapterLastPage>393</ChapterLastPage><ChapterCopyright><CopyrightHolderName>Springer-Verlag Berlin Heidelberg</CopyrightHolderName><CopyrightYear>2001</CopyrightYear></ChapterCopyright><ChapterHistory><RegistrationDate><Year>2001</Year><Month>8</Month><Day>29</Day></RegistrationDate><OnlineDate><Year>2001</Year><Month>8</Month><Day>30</Day></OnlineDate></ChapterHistory><ChapterGrants Type="Regular"><MetadataGrant Grant="OpenAccess"></MetadataGrant><AbstractGrant Grant="OpenAccess"></AbstractGrant><BodyPDFGrant Grant="Restricted"></BodyPDFGrant><BodyHTMLGrant Grant="Restricted"></BodyHTMLGrant><BibliographyGrant Grant="Restricted"></BibliographyGrant><ESMGrant Grant="Restricted"></ESMGrant></ChapterGrants><ChapterContext><SeriesID>558</SeriesID><PartID>1</PartID><BookID>3-540-44795-4</BookID><BookTitle>Machine Learning: ECML 2001</BookTitle></ChapterContext></ChapterInfo><ChapterHeader><AuthorGroup><Author AffiliationIDS="Aff5 Aff6"><AuthorName DisplayOrder="Western"><GivenName>Ann</GivenName><FamilyName>Nowé</FamilyName></AuthorName><Contact><Email>asnowe@info.vub.ac.be</Email></Contact></Author><Author AffiliationIDS="Aff5 Aff6"><AuthorName DisplayOrder="Western"><GivenName>Johan</GivenName><FamilyName>Parent</FamilyName></AuthorName><Contact><Email>johan@info.vub.ac.be</Email></Contact></Author><Author AffiliationIDS="Aff5 Aff6"><AuthorName DisplayOrder="Western"><GivenName>Katja</GivenName><FamilyName>Verbeeck</FamilyName></AuthorName><Contact><Email>kaverbee@vub.ac.be</Email></Contact></Author><Affiliation ID="Aff5"><OrgName>Computational Modeling Lab</OrgName><OrgAddress><Country>Belgium</Country></OrgAddress></Affiliation><Affiliation ID="Aff6"><OrgName>Vrije Universiteit Brussel</OrgName><OrgAddress><Country>Belgium</Country></OrgAddress></Affiliation></AuthorGroup><Abstract ID="Abs1" Language="En"><Heading>Abstract</Heading><Para>Coordination is an important issue in multiagent systems. Within the stochastic game framework this problem translates to policy learning in a joint action space. This technique however suffers some important drawbacks like the assumption of the existence of a unique Nash equilibrium and synchronicity, the need for central control, the cost of communication, etc. Moreover in general sum games it is not always clear which policies should be learned. Playing pure Nash equilibrium is often unfair to at least one of the players, while playing a mixed strategy doesn’t give any guarantee for coordination and usually results in a sub-optimal payoff for all agents. In this work we show the usefulness of periodical policies, which arise as a side effect of the fairness conditions used by the agents. We are interested in games which assume competition between the players, but where the overall performance can only be as good as the performance of the poorest player. Players are social distributed reinforcement learners, who have to learn to equalize their payoff. Our approach is illustrated on synchronous one-step games as well as on asynchronous job scheduling games.</Para></Abstract></ChapterHeader><NoBody></NoBody></Chapter></Part></Book><SubSeries><SubSeriesInfo><SubSeriesID>1244</SubSeriesID><SubSeriesTitle Language="En">Lecture Notes in Artificial Intelligence</SubSeriesTitle><SubSeriesSubTitle Language="En">Subseries of Lecture Notes in Computer Science</SubSeriesSubTitle></SubSeriesInfo><SubSeriesHeader><EditorGroup><Editor AffiliationIDS="Aff3"><EditorName DisplayOrder="Western"><GivenName>Jaime</GivenName><GivenName>G.</GivenName><FamilyName>Carbonell</FamilyName></EditorName></Editor><Editor AffiliationIDS="Aff4"><EditorName DisplayOrder="Western"><GivenName>Jörg</GivenName><FamilyName>Siekmann</FamilyName></EditorName></Editor><Affiliation ID="Aff3"><OrgName>Carnegie Mellon University</OrgName><OrgAddress><City>Pittsburgh</City><State>PA</State><Country>USA</Country></OrgAddress></Affiliation><Affiliation ID="Aff4"><OrgName>University of Saarland</OrgName><OrgAddress><City>Saarbrücken</City><Country>Germany</Country></OrgAddress></Affiliation></EditorGroup></SubSeriesHeader></SubSeries></Series></Publisher></istex:document></istex:metadataXml><mods version="3.6"><titleInfo lang="en"><title>Social Agents Playing a Periodical Policy</title></titleInfo><titleInfo type="alternative" contentType="CDATA" lang="en"><title>Social Agents Playing a Periodical Policy</title></titleInfo><name type="personal"><namePart type="given">Ann</namePart><namePart type="family">Nowé</namePart><affiliation>Computational Modeling Lab, Belgium</affiliation><affiliation>Vrije Universiteit Brussel, Belgium</affiliation><affiliation>E-mail: asnowe@info.vub.ac.be</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">Johan</namePart><namePart type="family">Parent</namePart><affiliation>Computational Modeling Lab, Belgium</affiliation><affiliation>Vrije Universiteit Brussel, Belgium</affiliation><affiliation>E-mail: johan@info.vub.ac.be</affiliation><role><roleTerm type="text">author</roleTerm></role></name><name type="personal"><namePart type="given">Katja</namePart><namePart type="family">Verbeeck</namePart><affiliation>Computational Modeling Lab, Belgium</affiliation><affiliation>Vrije Universiteit Brussel, Belgium</affiliation><affiliation>E-mail: kaverbee@vub.ac.be</affiliation><role><roleTerm type="text">author</roleTerm></role></name><typeOfResource>text</typeOfResource><genre type="Book Chapter">OriginalPaper</genre><originInfo><publisher>Springer Berlin Heidelberg</publisher><place><placeTerm type="text">Berlin, Heidelberg</placeTerm></place><copyrightDate encoding="w3cdtf">2001</copyrightDate></originInfo><language><languageTerm type="code" authority="rfc3066">en</languageTerm><languageTerm type="code" authority="iso639-2b">eng</languageTerm></language><physicalDescription><internetMediaType>text/html</internetMediaType></physicalDescription><abstract lang="en">Abstract: Coordination is an important issue in multiagent systems. Within the stochastic game framework this problem translates to policy learning in a joint action space. This technique however suffers some important drawbacks like the assumption of the existence of a unique Nash equilibrium and synchronicity, the need for central control, the cost of communication, etc. Moreover in general sum games it is not always clear which policies should be learned. Playing pure Nash equilibrium is often unfair to at least one of the players, while playing a mixed strategy doesn’t give any guarantee for coordination and usually results in a sub-optimal payoff for all agents. In this work we show the usefulness of periodical policies, which arise as a side effect of the fairness conditions used by the agents. We are interested in games which assume competition between the players, but where the overall performance can only be as good as the performance of the poorest player. Players are social distributed reinforcement learners, who have to learn to equalize their payoff. Our approach is illustrated on synchronous one-step games as well as on asynchronous job scheduling games.</abstract><relatedItem type="host"><titleInfo><title>Machine Learning: ECML 2001</title><subTitle>12th European Conference on Machine Learning Freiburg, Germany, September 5–7, 2001 Proceedings</subTitle></titleInfo><name type="personal"><namePart type="given">Luc</namePart><namePart type="family">De Raedt</namePart><affiliation>Department of Computer Science, Albert-Ludwigs University Freiburg, Georges Köhler-Allee, Geb. 079, 79110, Freiburg, Germany</affiliation><affiliation>E-mail: deraedt@informatik.uni-freiburg.de</affiliation><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">Peter</namePart><namePart type="family">Flach</namePart><affiliation>Department of Computer Science, University of Bristol, Merchant Ventures Bldg., Woodland Road, BS8 1UB, Bristol, UK</affiliation><affiliation>E-mail: peter.flach@bristol.ac.uk</affiliation><role><roleTerm type="text">editor</roleTerm></role></name><genre type="book series">Proceedings</genre><originInfo><copyrightDate encoding="w3cdtf">2001</copyrightDate><issuance>monographic</issuance></originInfo><subject><genre>Book Subject Collection</genre><topic authority="SpringerSubjectCodes" authorityURI="SUCO11645">Computer Science</topic></subject><subject><genre>Book Subject Group</genre><topic authority="SpringerSubjectCodes" authorityURI="I">Computer Science</topic><topic authority="SpringerSubjectCodes" authorityURI="I21017">Artificial Intelligence (incl. Robotics)</topic><topic authority="SpringerSubjectCodes" authorityURI="I16021">Algorithm Analysis and Problem Complexity</topic><topic authority="SpringerSubjectCodes" authorityURI="I16048">Mathematical Logic and Formal Languages</topic></subject><identifier type="DOI">10.1007/3-540-44795-4</identifier><identifier type="ISBN">978-3-540-42536-6</identifier><identifier type="eISBN">978-3-540-44795-5</identifier><identifier type="BookTitleID">69457</identifier><identifier type="BookID">3-540-44795-4</identifier><identifier type="BookChapterCount">55</identifier><identifier type="BookVolumeNumber">2167</identifier><identifier type="BookSequenceNumber">2167</identifier><identifier type="PartChapterCount">50</identifier><part><date>2001</date><detail type="part"><title>Regular Papers</title></detail><detail type="volume"><number>2167</number><caption>vol.</caption></detail><extent unit="pages"><start>382</start><end>393</end></extent></part><recordInfo><recordOrigin>Springer-Verlag Berlin Heidelberg, 2001</recordOrigin></recordInfo></relatedItem><relatedItem type="series"><titleInfo><title>Lecture Notes in Computer Science</title></titleInfo><name type="personal"><namePart type="given">G.</namePart><namePart type="family">Goos</namePart><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">J.</namePart><namePart type="family">Hartmanis</namePart><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">J.</namePart><namePart type="family">van Leeuwen</namePart><role><roleTerm type="text">editor</roleTerm></role></name><genre type="book series">series</genre><originInfo><copyrightDate encoding="w3cdtf">2001</copyrightDate><issuance>serial</issuance></originInfo><relatedItem type="constituent"><titleInfo><title>Lecture Notes in Artificial Intelligence</title><subTitle>Subseries of Lecture Notes in Computer Science</subTitle></titleInfo><name type="personal"><namePart type="given">G.</namePart><namePart type="family">Goos</namePart><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">J.</namePart><namePart type="family">Hartmanis</namePart><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">J.</namePart><namePart type="family">van Leeuwen</namePart><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">Luc</namePart><namePart type="family">De Raedt</namePart><affiliation>Department of Computer Science, Albert-Ludwigs University Freiburg, Georges Köhler-Allee, Geb. 079, 79110, Freiburg, Germany</affiliation><affiliation>E-mail: deraedt@informatik.uni-freiburg.de</affiliation><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">Peter</namePart><namePart type="family">Flach</namePart><affiliation>Department of Computer Science, University of Bristol, Merchant Ventures Bldg., Woodland Road, BS8 1UB, Bristol, UK</affiliation><affiliation>E-mail: peter.flach@bristol.ac.uk</affiliation><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">Jaime</namePart><namePart type="given">G.</namePart><namePart type="family">Carbonell</namePart><affiliation>Carnegie Mellon University, Pittsburgh, PA, USA</affiliation><role><roleTerm type="text">editor</roleTerm></role></name><name type="personal"><namePart type="given">Jörg</namePart><namePart type="family">Siekmann</namePart><affiliation>University of Saarland, Saarbrücken, Germany</affiliation><role><roleTerm type="text">editor</roleTerm></role></name><genre>sub-series</genre><identifier type="SubSeriesID">1244</identifier></relatedItem><identifier type="ISSN">0302-9743</identifier><identifier type="SeriesID">558</identifier><recordInfo><recordOrigin>Springer-Verlag Berlin Heidelberg, 2001</recordOrigin></recordInfo></relatedItem><identifier type="istex">B9569E9E27EC810019E7B082F8F573E317E9C04A</identifier><identifier type="DOI">10.1007/3-540-44795-4_33</identifier><identifier type="ChapterID">Chap33</identifier><identifier type="ChapterID">33</identifier><accessCondition type="use and reproduction" contentType="Copyright">Springer-Verlag Berlin Heidelberg</accessCondition><recordInfo><recordOrigin>Springer-Verlag Berlin Heidelberg, 2001</recordOrigin></recordInfo></mods></metadata><enrichments><istex:refBibTEI uri="https://api.istex.fr/document/B9569E9E27EC810019E7B082F8F573E317E9C04A/enrichments/refBib"><teiHeader></teiHeader><text><front></front><body></body><back><listBibl><biblStruct xml:id="b0"><analytic><title level="a" type="main">The dynamics of reinforcement learning in cooperative multi-agent systems</title><author><persName><forename type="first">Claus</forename><forename type="middle">C</forename><surname>Boutilier</surname></persName></author><author><persName><forename type="first">C</forename></persName></author></analytic><monogr><title level="m">Proceedings of the fifteenth National Conference on Artificial Intelligence</title><meeting>the fifteenth National Conference on Artificial Intelligence</meeting><imprint><date type="published" when="1998"></date><biblScope unit="page" from="746" to="752"></biblScope></imprint></monogr></biblStruct><biblStruct xml:id="b1"><monogr><title level="m" type="main">Game Theory Evolving: A Problem-Centered Introduction to Modeling Strategic Behavior</title><author><persName><forename type="first">H</forename><surname>Gintis</surname></persName></author><imprint><date type="published" when="2000"></date><publisher>Princeton University Press</publisher></imprint></monogr></biblStruct><biblStruct xml:id="b2"><analytic><title level="a" type="main">A Theory of Fairness, Competition, and Cooperation</title><author><persName><forename type="first">E</forename><surname>Fehr</surname></persName></author><author><persName><forename type="first">K</forename><forename type="middle">M</forename><surname>Schmidt</surname></persName></author></analytic><monogr><title level="j">Quarterly Journal of Economics</title><imprint><biblScope unit="volume">114</biblScope><biblScope unit="page" from="817" to="868"></biblScope><date type="published" when="1999"></date></imprint></monogr></biblStruct><biblStruct xml:id="b3"><analytic><title level="a" type="main">The tragedy of the commons</title><author><persName><forename type="first">G</forename><surname>Hardin</surname></persName></author></analytic><monogr><title level="j">Science</title><imprint><biblScope unit="volume">162</biblScope><biblScope unit="page" from="1243" to="1248"></biblScope><date type="published" when="1968"></date></imprint></monogr></biblStruct><biblStruct xml:id="b4"><analytic><title></title><author><persName><forename type="first">J</forename><surname>Hu</surname></persName></author><author><persName><forename type="first">M</forename><forename type="middle">P</forename><surname>Wellman</surname></persName></author></analytic><monogr><title level="m">Multi Agent Reinforcement Learning in Stochastic Games</title><imprint><date type="published" when="1999"></date></imprint></monogr><note>Submitted</note></biblStruct><biblStruct xml:id="b5"><analytic><title level="a" type="main">Markov games as a framework for multi-agent reinforcement learning</title><author><persName><forename type="first">M</forename><forename type="middle">L</forename><surname>Litmann</surname></persName></author></analytic><monogr><title level="m">Proceedings of the Eleventh International Conference on Machine Learning</title><meeting>the Eleventh International Conference on Machine Learning</meeting><imprint><date type="published" when="1994"></date><biblScope unit="page" from="157" to="163"></biblScope></imprint></monogr></biblStruct><biblStruct xml:id="b6"><analytic><title level="a" type="main">Maximum Games, Dominance Solvability, and Coordination</title><author><persName><forename type="first">M</forename><surname>Mariotti</surname></persName></author></analytic><monogr><title level="j">Games and Economic Behavior</title><imprint><biblScope unit="volume">31</biblScope><biblScope unit="page" from="97" to="105"></biblScope><date type="published" when="2000"></date></imprint></monogr></biblStruct><biblStruct xml:id="b7"><analytic><title></title><author><persName><forename type="first">K</forename><surname>Narendra</surname></persName></author><author><persName><forename type="first">M</forename><surname>Thathachar</surname></persName></author></analytic><monogr><title level="m">Learning Automata: An Introduction</title><imprint><publisher>Prentice-Hall</publisher><date type="published" when="1989"></date></imprint></monogr></biblStruct><biblStruct xml:id="b8"><analytic><title level="a" type="main">Distributed Reinforcement learning, Loadbased Routing a case study</title><author><persName><forename type="first">A</forename><surname>Nowé</surname></persName></author><author><persName><forename type="first">K</forename><surname>Verbeeck</surname></persName></author></analytic><monogr><title level="m">Proceedings of the Neural, Symbolic and Reinforcement Methods for sequence Learning Workshop at ijcai99</title><meeting>the Neural, Symbolic and Reinforcement Methods for sequence Learning Workshop at ijcai99</meeting><imprint></imprint></monogr></biblStruct><biblStruct xml:id="b9"><analytic><title></title><author><persName><forename type="first">A</forename><surname>Nowé</surname></persName></author><author><persName><forename type="first">K</forename><surname>Verbeeck</surname></persName></author><author><persName><forename type="first">T</forename><surname>Lenaerts</surname></persName></author></analytic><monogr><title level="j">Homo Egualis Society Technical report Computational Modeling Lab. Maart Vrije Uniuversiteit Brussel</title><imprint><date type="published" when="2001"></date></imprint></monogr></biblStruct><biblStruct xml:id="b10"><analytic><title></title><author><persName><forename type="first">Osborne</forename><forename type="middle">J O</forename><surname>Rubinstein</surname></persName></author><author><persName><forename type="first">A</forename></persName></author></analytic><monogr><title level="m">A course in game theory</title><meeting><address><addrLine>Cambridge, MA</addrLine></address></meeting><imprint><publisher>MIT Press</publisher><date type="published" when="1994"></date></imprint></monogr></biblStruct><biblStruct xml:id="b11"><analytic><title></title><author><persName><forename type="first">Smith</forename><forename type="middle">J M</forename></persName></author></analytic><monogr><title level="m">Evolution and the Theory of Games</title><imprint><publisher>Cambridge University Press</publisher><date type="published" when="1982"></date></imprint></monogr></biblStruct><biblStruct xml:id="b12"><monogr><title level="m" type="main">Reinforcement Learning: An introduction</title><author><persName><forename type="first">R</forename><forename type="middle">S</forename><surname>Sutton</surname></persName></author><author><persName><forename type="first">A</forename><forename type="middle">G</forename><surname>Barto</surname></persName></author><imprint><date type="published" when="1998"></date><publisher>MIT Press</publisher><pubPlace>Cambridge, MA</pubPlace></imprint></monogr></biblStruct></listBibl></back></text></istex:refBibTEI></enrichments></istex></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Musique/explor/MozartV1/Data/Istex/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 001674 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Corpus/biblio.hfd -nk 001674 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Musique
|area= MozartV1
|flux= Istex
|étape= Corpus
|type= RBID
|clé= ISTEX:B9569E9E27EC810019E7B082F8F573E317E9C04A
|texte= Social Agents Playing a Periodical Policy
}}
| This area was generated with Dilib version V0.6.20. |  |