Bayesian online algorithms for learning in discrete Hidden Markov Models

dc.creatorALAMINO, Roberto C.
dc.creatorCaticha, Nestor
dc.date.accessioned2012-04-18T23:58:07Z
dc.date.accessioned2018-07-04T14:40:28Z
dc.date.available2012-04-18T23:58:07Z
dc.date.available2018-07-04T14:40:28Z
dc.date.created2012-04-18T23:58:07Z
dc.date.issued2008
dc.identifierDISCRETE AND CONTINUOUS DYNAMICAL SYSTEMS-SERIES B, v.9, n.1, p.1-10, 2008
dc.identifier1531-3492
dc.identifierhttp://producao.usp.br/handle/BDPI/16125
dc.identifierhttp://aimsciences.org/journals/pdfs.jsp?paperID=2980&mode=full
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/1612947
dc.description.abstractWe propose and analyze two different Bayesian online algorithms for learning in discrete Hidden Markov Models and compare their performance with the already known Baldi-Chauvin Algorithm. Using the Kullback-Leibler divergence as a measure of generalization we draw learning curves in simplified situations for these algorithms and compare their performances.
dc.languageeng
dc.publisherAMER INST MATHEMATICAL SCIENCES
dc.relationDiscrete and Continuous Dynamical Systems-series B
dc.rightsCopyright AMER INST MATHEMATICAL SCIENCES
dc.rightsopenAccess
dc.subjectHMM
dc.subjectonline algorithm
dc.subjectgeneralization error
dc.subjectBayesian algorithm
dc.titleBayesian online algorithms for learning in discrete Hidden Markov Models
dc.typeArtículos de revistas


Este ítem pertenece a la siguiente institución