Ítem
Solo Metadatos

Time Series Forecasting using Recurrent Neural Networks modified by Bayesian Inference in the Learning Process

dc.creatorRivero C.R.spa
dc.creatorPucheta J.spa
dc.creatorOtaño P.spa
dc.creatorOrjuela Cañón, Alvaro Davidspa
dc.creatorPatiño D.spa
dc.creatorFranco L.spa
dc.creatorGorrostieta E.spa
dc.creatorPuglisi J.L.spa
dc.creatorJuarez G.spa
dc.date.accessioned2020-05-25T23:56:10Z
dc.date.available2020-05-25T23:56:10Z
dc.date.created2019spa
dc.description.abstractTypically, time series forecasting is done by using models based directly on the past observations from the same sequence. In these cases, when the model is learning from data, there is not an extra quantity of noiseless data available and computational resources are unlimited. In practice, it is necessary to deal with finite noisy datasets, which lead to uncertainty about what so appropriate the model is. For this, the employment of models based on Bayesian inference are preferable. Then, probabilities are treated as a way to represent the subjective uncertainty from rational agent, performing an approximated inference by maximizing a lower bound on the marginal likelihood. A modified algorithm using long-short memory recurrent neural networks for time series forecasting was presented. This new approach was chosen in order to be as close as possible to the original series in the sense of minimizing the associated Kullback-Leibler Information Criterion. A simulation study was conducted to evaluate and illustrate results, comparing this approach with Bayesian neural-networks-based algorithms for artificial chaotic time-series. © 2019 IEEE.eng
dc.format.mimetypeapplication/pdf
dc.identifier.doihttps://doi.org/10.1109/ColCACI.2019.8781984
dc.identifier.urihttps://repository.urosario.edu.co/handle/10336/22344
dc.language.isoengspa
dc.publisherInstitute of Electrical and Electronics Engineers Inc.spa
dc.relation.citationTitle2019 IEEE Colombian Conference on Applications in Computational Intelligence ColCACI 2019 - Proceedings
dc.relation.ispartof2019 IEEE Colombian Conference on Applications in Computational Intelligence, ColCACI 2019 - Proceedings,(2019)spa
dc.relation.urihttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85070921034&doi=10.1109%2fColCACI.2019.8781984&partnerID=40&md5=d2cd3fc8618b4503c08039aed3d6c148spa
dc.rights.accesRightsinfo:eu-repo/semantics/openAccess
dc.rights.accesoAbierto (Texto Completo)spa
dc.source.instnameinstname:Universidad del Rosariospa
dc.source.reponamereponame:Repositorio Institucional EdocURspa
dc.subject.keywordBayesian networksspa
dc.subject.keywordForecastingspa
dc.subject.keywordInference enginesspa
dc.subject.keywordLearning systemsspa
dc.subject.keywordTime seriesspa
dc.subject.keywordBayesianspa
dc.subject.keywordBayesian neural networksspa
dc.subject.keywordComputational resourcesspa
dc.subject.keywordKullback Leibler divergencespa
dc.subject.keywordKullback-Leibler informationspa
dc.subject.keywordMarginal likelihoodspa
dc.subject.keywordSubjective uncertaintyspa
dc.subject.keywordTime series forecastingspa
dc.subject.keywordRecurrent neural networksspa
dc.subject.keywordBayesian approximationspa
dc.subject.keywordKullback-Leibler Divergencespa
dc.subject.keywordRecurrent neural networkspa
dc.subject.keywordTime Series Forecastingspa
dc.titleTime Series Forecasting using Recurrent Neural Networks modified by Bayesian Inference in the Learning Processspa
dc.typeconferenceObjecteng
dc.type.hasVersioninfo:eu-repo/semantics/publishedVersion
dc.type.spaDocumento de conferenciaspa
Archivos
Colecciones