Heuristically-accelerated multiagent reinforcement learning

Página do item simplificado
dc.contributor.authorReinaldo Bianchi
dc.contributor.authorMARTINS, M. F.
dc.contributor.authorRIBEIRO, C. H. C.
dc.contributor.authorCOSTA, A. H. R.
dc.contributor.authorOrcidhttps://orcid.org/0000-0001-9097-827X
dc.date.accessioned2022-01-12T22:00:41Z
dc.date.available2022-01-12T22:00:41Z
dc.date.issued2014-02-05
dc.description.abstractThis paper presents a novel class of algorithms, called Heuristically-Accelerated Multiagent Reinforcement Learning (HAMRL), which allows the use of heuristics to speed up well-known multiagent reinforcement learning (RL) algorithms such as the Minimax-Q. Such HAMRL algorithms are characterized by a heuristic function, which suggests the selection of particular actions over others. This function represents an initial action selection policy, which can be handcrafted, extracted from previous experience in distinct domains, or learnt from observation. To validate the proposal, a thorough theoretical analysis proving the convergence of four algorithms from the HAMRL class (HAMMQ, HAMQ}(λ, HAMQS, and HAMS) is presented. In addition, a comprehensive systematical evaluation was conducted in two distinct adversarial domains. The results show that even the most straightforward heuristics can produce virtually optimal action selection policies in much fewer episodes, significantly improving the performance of the HAMRL over vanilla RL algorithms. © 2013 IEEE.
dc.description.firstpage252
dc.description.issuenumber2
dc.description.lastpage265
dc.description.volume44
dc.identifier.citationBIANCHI, R.; MARTINS, M. F.; RIBEIRO, C. H. C.; COSTA, A. H. R. Heuristically-accelerated multiagent reinforcement learning. IEEE Transactions on Cybernetics, v. 44, n. 2, p. 252-265, Feb. 2014.
dc.identifier.doi10.1109/TCYB.2013.2253094
dc.identifier.issn2168-2267
dc.identifier.urihttps://repositorio.fei.edu.br/handle/FEI/4016
dc.relation.ispartofIEEE Transactions on Cybernetics
dc.rightsAcesso Restrito
dc.subject.otherlanguageArtificial intelligence
dc.subject.otherlanguageheuristic algorithms
dc.subject.otherlanguagemachine learning
dc.subject.otherlanguagemultiagent systems
dc.titleHeuristically-accelerated multiagent reinforcement learning
dc.typeArtigo
fei.scopus.citations58
fei.scopus.eid2-s2.0-84893355297
fei.scopus.subjectAction selection
fei.scopus.subjectHeuristic functions
fei.scopus.subjectMulti-agent reinforcement learning
fei.scopus.subjectOptimal actions
fei.scopus.subjectSpeed up
fei.scopus.updated2024-08-01
fei.scopus.urlhttps://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84893355297&origin=inward
Arquivos
Coleções
Artigos