Transferring knowledge as heuristics in reinforcement learning: A case-based approach
dc.contributor.author | Bianchi R.A.C. | |
dc.contributor.author | Celiberto L.A. | |
dc.contributor.author | Santos P.E. | |
dc.contributor.author | Matsuura J.P. | |
dc.contributor.author | Lopez De Mantaras R. | |
dc.date.accessioned | 2019-08-19T23:45:20Z | |
dc.date.available | 2019-08-19T23:45:20Z | |
dc.date.issued | 2015 | |
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.firstpage | 102 | |
dc.description.lastpage | 121 | |
dc.description.volume | 226 | |
dc.identifier.citation | BIANCHI, 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.doi | 10.1016/j.artint.2015.05.008 | |
dc.identifier.issn | 0004-3702 | |
dc.identifier.uri | https://repositorio.fei.edu.br/handle/FEI/1217 | |
dc.relation.ispartof | Artificial Intelligence | |
dc.rights | Acesso Aberto | |
dc.subject.otherlanguage | Case-based reasoning | |
dc.subject.otherlanguage | Reinforcement learning | |
dc.subject.otherlanguage | Transfer learning | |
dc.title | Transferring knowledge as heuristics in reinforcement learning: A case-based approach | |
dc.type | Artigo | |
fei.scopus.citations | 59 | |
fei.scopus.eid | 2-s2.0-84930960233 | |
fei.scopus.subject | 2D simulations | |
fei.scopus.subject | Case-based approach | |
fei.scopus.subject | Empirical evaluations | |
fei.scopus.subject | Humanoid robot | |
fei.scopus.subject | Learning process | |
fei.scopus.subject | Meta-algorithms | |
fei.scopus.subject | Target domain | |
fei.scopus.subject | Transfer learning | |
fei.scopus.updated | 2024-03-04 | |
fei.scopus.url | https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84930960233&origin=inward |