Contribution to the Control of a MAS’s Global Behaviour: Reinforcement Learning Tools
Identifieur interne : 003976 ( Main/Exploration ); précédent : 003975; suivant : 003977Contribution to the Control of a MAS’s Global Behaviour: Reinforcement Learning Tools
Auteurs : François Klein [France] ; Christine Bourjot [France] ; Vincent Chevrier [France]Source :
- Lecture Notes in Computer Science [ 0302-9743 ]
English descriptors
Abstract
Abstract: Reactive multi-agent systems present global behaviours uneasily linked to their local dynamics. When it comes to controlling such a system, usual analytical tools are difficult to use so specific techniques have to be engineered. We propose an experimental dynamical approach to enhance the control of the global behaviour of a reactive multi-agent system. We use reinforcement learning tools to link global information of the system to control actions. We propose to use the behaviour of the system as this global information. The behaviour of the whole system is controlled thanks to actions at different levels instead of building the behaviours of the agents, so that the complexity of the approach does not directly depend on the number of agents. The controllability is evaluated in terms of rate of convergence towards a target behaviour. We compare the results obtained on a toy example with the usual approach of parameter setting.
Url:
DOI: 10.1007/978-3-642-02562-4_10
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 003491
- to stream Istex, to step Curation: 003449
- to stream Istex, to step Checkpoint: 000A83
- to stream Hal, to step Corpus: 001903
- to stream Hal, to step Curation: 001903
- to stream Hal, to step Checkpoint: 002F77
- to stream Main, to step Merge: 003A54
- to stream Main, to step Curation: 003976
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Contribution to the Control of a MAS’s Global Behaviour: Reinforcement Learning Tools</title>
<author><name sortKey="Klein, Francois" sort="Klein, Francois" uniqKey="Klein F" first="François" last="Klein">François Klein</name>
</author>
<author><name sortKey="Bourjot, Christine" sort="Bourjot, Christine" uniqKey="Bourjot C" first="Christine" last="Bourjot">Christine Bourjot</name>
</author>
<author><name sortKey="Chevrier, Vincent" sort="Chevrier, Vincent" uniqKey="Chevrier V" first="Vincent" last="Chevrier">Vincent Chevrier</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:DDFCDA2D41606A621F882FFC3BDB100DF8D07B23</idno>
<date when="2009" year="2009">2009</date>
<idno type="doi">10.1007/978-3-642-02562-4_10</idno>
<idno type="url">https://api.istex.fr/ark:/67375/HCB-2WP12GMJ-3/fulltext.pdf</idno>
<idno type="wicri:Area/Istex/Corpus">003491</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">003491</idno>
<idno type="wicri:Area/Istex/Curation">003449</idno>
<idno type="wicri:Area/Istex/Checkpoint">000A83</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000A83</idno>
<idno type="wicri:doubleKey">0302-9743:2009:Klein F:contribution:to:the</idno>
<idno type="wicri:source">HAL</idno>
<idno type="RBID">Hal:inria-00400348</idno>
<idno type="url">https://hal.inria.fr/inria-00400348</idno>
<idno type="wicri:Area/Hal/Corpus">001903</idno>
<idno type="wicri:Area/Hal/Curation">001903</idno>
<idno type="wicri:Area/Hal/Checkpoint">002F77</idno>
<idno type="wicri:explorRef" wicri:stream="Hal" wicri:step="Checkpoint">002F77</idno>
<idno type="wicri:Area/Main/Merge">003A54</idno>
<idno type="wicri:Area/Main/Curation">003976</idno>
<idno type="wicri:Area/Main/Exploration">003976</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Contribution to the Control of a MAS’s Global Behaviour: Reinforcement Learning Tools</title>
<author><name sortKey="Klein, Francois" sort="Klein, Francois" uniqKey="Klein F" first="François" last="Klein">François Klein</name>
<affiliation></affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">France</country>
</affiliation>
</author>
<author><name sortKey="Bourjot, Christine" sort="Bourjot, Christine" uniqKey="Bourjot C" first="Christine" last="Bourjot">Christine Bourjot</name>
<affiliation></affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">France</country>
</affiliation>
</author>
<author><name sortKey="Chevrier, Vincent" sort="Chevrier, Vincent" uniqKey="Chevrier V" first="Vincent" last="Chevrier">Vincent Chevrier</name>
<affiliation></affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">France</country>
</affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="s" type="main" xml:lang="en">Lecture Notes in Computer Science</title>
<idno type="ISSN">0302-9743</idno>
<idno type="eISSN">1611-3349</idno>
<idno type="ISSN">0302-9743</idno>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><idno type="ISSN">0302-9743</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="mix" xml:lang="en"><term>Control</term>
<term>MAS</term>
<term>emergence</term>
<term>experimental approach</term>
<term>global behaviour</term>
<term>reinforcement learning</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Abstract: Reactive multi-agent systems present global behaviours uneasily linked to their local dynamics. When it comes to controlling such a system, usual analytical tools are difficult to use so specific techniques have to be engineered. We propose an experimental dynamical approach to enhance the control of the global behaviour of a reactive multi-agent system. We use reinforcement learning tools to link global information of the system to control actions. We propose to use the behaviour of the system as this global information. The behaviour of the whole system is controlled thanks to actions at different levels instead of building the behaviours of the agents, so that the complexity of the approach does not directly depend on the number of agents. The controllability is evaluated in terms of rate of convergence towards a target behaviour. We compare the results obtained on a toy example with the usual approach of parameter setting.</div>
</front>
</TEI>
<affiliations><list><country><li>France</li>
</country>
</list>
<tree><country name="France"><noRegion><name sortKey="Klein, Francois" sort="Klein, Francois" uniqKey="Klein F" first="François" last="Klein">François Klein</name>
</noRegion>
<name sortKey="Bourjot, Christine" sort="Bourjot, Christine" uniqKey="Bourjot C" first="Christine" last="Bourjot">Christine Bourjot</name>
<name sortKey="Chevrier, Vincent" sort="Chevrier, Vincent" uniqKey="Chevrier V" first="Vincent" last="Chevrier">Vincent Chevrier</name>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 003976 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 003976 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Main |étape= Exploration |type= RBID |clé= ISTEX:DDFCDA2D41606A621F882FFC3BDB100DF8D07B23 |texte= Contribution to the Control of a MAS’s Global Behaviour: Reinforcement Learning Tools }}
This area was generated with Dilib version V0.6.33. |