A Model-Based Actor-Critic Algorithm in Continuous Time and Space
Identifieur interne : 003860 ( Crin/Corpus ); précédent : 003859; suivant : 003861A Model-Based Actor-Critic Algorithm in Continuous Time and Space
Auteurs : Rémi CoulomSource :
English descriptors
Abstract
This paper presents a model-based actor-critic algorithm in continuous time and space. Two function approximators are used : one learns the policy (the actor) and the other learns the state-value function (the critic). The critic learns with the TD(lambda) algorithm and the actor by gradient ascent on the Hamiltonian. A similar algorithm had been proposed by Doya, but this one is more general. This algorithm was applied successfully to teach simulated articulated robots to swim.
Links to Exploration step
CRIN:coulom03aLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="340">A Model-Based Actor-Critic Algorithm in Continuous Time and Space</title>
</titleStmt>
<publicationStmt><idno type="RBID">CRIN:coulom03a</idno>
<date when="2003" year="2003">2003</date>
<idno type="wicri:Area/Crin/Corpus">003860</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">A Model-Based Actor-Critic Algorithm in Continuous Time and Space</title>
<author><name sortKey="Coulom, Remi" sort="Coulom, Remi" uniqKey="Coulom R" first="Rémi" last="Coulom">Rémi Coulom</name>
</author>
</analytic>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>dynamic programming</term>
<term>optimal control</term>
<term>reinforcement learning</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en" wicri:score="1727">This paper presents a model-based actor-critic algorithm in continuous time and space. Two function approximators are used : one learns the policy (the actor) and the other learns the state-value function (the critic). The critic learns with the TD(lambda) algorithm and the actor by gradient ascent on the Hamiltonian. A similar algorithm had been proposed by Doya, but this one is more general. This algorithm was applied successfully to teach simulated articulated robots to swim.</div>
</front>
</TEI>
<BibTex type="inproceedings"><ref>coulom03a</ref>
<crinnumber>A03-R-125</crinnumber>
<category>3</category>
<equipe>CORTEX</equipe>
<author><e>Coulom, Rémi</e>
</author>
<title>A Model-Based Actor-Critic Algorithm in Continuous Time and Space</title>
<booktitle>{Sixth European Workshop on Reinforcement Learning - EWRL6, Nancy, France}</booktitle>
<year>2003</year>
<month>Sep</month>
<url>http://www.loria.fr/publications/2003/A03-R-125/A03-R-125.ps</url>
<keywords><e>reinforcement learning</e>
<e>optimal control</e>
<e>dynamic programming</e>
</keywords>
<abstract>This paper presents a model-based actor-critic algorithm in continuous time and space. Two function approximators are used : one learns the policy (the actor) and the other learns the state-value function (the critic). The critic learns with the TD(lambda) algorithm and the actor by gradient ascent on the Hamiltonian. A similar algorithm had been proposed by Doya, but this one is more general. This algorithm was applied successfully to teach simulated articulated robots to swim.</abstract>
</BibTex>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Crin/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 003860 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Crin/Corpus/biblio.hfd -nk 003860 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Crin |étape= Corpus |type= RBID |clé= CRIN:coulom03a |texte= A Model-Based Actor-Critic Algorithm in Continuous Time and Space }}
![]() | This area was generated with Dilib version V0.6.33. | ![]() |