Transferring knowledge as heuristics in reinforcement learning: A case-based approach

Página do item simplificado
dc.contributor.authorBianchi R.A.C.
dc.contributor.authorCeliberto L.A.
dc.contributor.authorSantos P.E.
dc.contributor.authorMatsuura J.P.
dc.contributor.authorLopez De Mantaras R.
dc.date.accessioned2019-08-19T23:45:20Z
dc.date.available2019-08-19T23:45:20Z
dc.date.issued2015
dc.description.abstract© 2015 Elsevier B.V.Abstract The goal of this paper is to propose and analyse a transfer learning meta-algorithm that allows the implementation of distinct methods using heuristics to accelerate a Reinforcement Learning procedure in one domain (the target) that are obtained from another (simpler) domain (the source domain). This meta-algorithm works in three stages: first, it uses a Reinforcement Learning step to learn a task on the source domain, storing the knowledge thus obtained in a case base; second, it does an unsupervised mapping of the source-domain actions to the target-domain actions; and, third, the case base obtained in the first stage is used as heuristics to speed up the learning process in the target domain. A set of empirical evaluations were conducted in two target domains: the 3D mountain car (using a learned case base from a 2D simulation) and stability learning for a humanoid robot in the Robocup 3D Soccer Simulator (that uses knowledge learned from the Acrobot domain). The results attest that our transfer learning algorithm outperforms recent heuristically-accelerated reinforcement learning and transfer learning algorithms.
dc.description.firstpage102
dc.description.lastpage121
dc.description.volume226
dc.identifier.citationBIANCHI, REINALDO A.C.; JUNIOR, LUIZ A. CELIBERTO; Santos, Paulo E.; MATSUURA, JACKSON P.; LÓPEZ DE MÀNTARAS, RAMÓN. Transferring knowledge as heuristics in reinforcement learning: a case-based approach. Artificial Intelligence (General Ed.), v. 226, p. 102-121, 2015.
dc.identifier.doi10.1016/j.artint.2015.05.008
dc.identifier.issn0004-3702
dc.identifier.urihttps://repositorio.fei.edu.br/handle/FEI/1217
dc.relation.ispartofArtificial Intelligence
dc.rightsAcesso Aberto
dc.subject.otherlanguageCase-based reasoning
dc.subject.otherlanguageReinforcement learning
dc.subject.otherlanguageTransfer learning
dc.titleTransferring knowledge as heuristics in reinforcement learning: A case-based approach
dc.typeArtigo
fei.scopus.citations59
fei.scopus.eid2-s2.0-84930960233
fei.scopus.subject2D simulations
fei.scopus.subjectCase-based approach
fei.scopus.subjectEmpirical evaluations
fei.scopus.subjectHumanoid robot
fei.scopus.subjectLearning process
fei.scopus.subjectMeta-algorithms
fei.scopus.subjectTarget domain
fei.scopus.subjectTransfer learning
fei.scopus.updated2024-03-04
fei.scopus.urlhttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84930960233&origin=inward
Arquivos
Coleções
Artigos