Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting

Autores
Rodríguez Rivero, Cristian; Pucheta, Julián; Patiño, Daniel; Puglisi, Jose Luis; Otaño, Paula; Franco, Leonardo; Juárez, Gustavo; Gorrostieta, Efrén; Orjuela-Cañon, Alvaro David
Año de publicación
2019
Idioma
inglés
Tipo de recurso
parte de libro
Estado
versión publicada
Descripción
Fil: Rodríguez Rivero, Cristian. Universidad Nacional de Córdoba; Argentina.
Fil: Pucheta, Julián. Universidad Nacional de Córdoba; Argentina.
Fil: Patiño, Daniel. Universidad Nacional de San Juan. Instituto de Automática; Argentina.
Fil: Puglisi, Jose Luis. Universidad de Carolina del Norte; Estados Unidos.
Fil: Otaño, Paula. Universidad Tecnológica Nacional. Facultad Regional Córdoba.
Fil: Franco, Leonardo. Universidad de Málaga; España.
Fil: Juárez, Gustavo. Universidad Nacional de Tucumán; Argentina.
Fil: Gorrostieta, Efrén. Universidad Autónoma de Querétaro; México.
Fil: Orjuela-Cañon, Alvaro David. Universidad del Rosario; Colombia.
For time series forecasting, obtaining models is based on the use of past observations from the same sequence. In those cases, when the model is learning from data, there is not an extra information that discuss about the quantity of noise inside the data available. In practice, it is necessary to deal with finite noisy datasets, which lead to uncertainty about the propriety of the model. For this problem, the employment of the Bayesian inference tools are preferable. A modified algorithm used for training a long-short term memory recurrent neural network for time series forecasting is presented. This approach was chosen to improve the forecasting of the original series, employing an implementation based on the minimization of the associated Kullback-Leibler Information Criterion. For comparison, a nonlinear autoregressive model implemented with a feedforward neural network was also presented. A simulation study was conducted to evaluate and illustrate results, comparing this approach with Bayesian neural-networks-based algorithms for artificial chaotic time-series and showing an improvement in terms of forecasting errors.
https://www.springer.com/gp/book/9783030362102?wt_mc=ThirdParty.SpringerLink.3.EPR653.About_eBook
publishedVersion
Fil: Rodríguez Rivero, Cristian. Universidad Nacional de Córdoba; Argentina.
Fil: Pucheta, Julián. Universidad Nacional de Córdoba; Argentina.
Fil: Patiño, Daniel. Universidad Nacional de San Juan. Instituto de Automática; Argentina.
Fil: Puglisi, Jose Luis. Universidad de Carolina del Norte; Estados Unidos.
Fil: Otaño, Paula. Universidad Tecnológica Nacional. Facultad Regional Córdoba.
Fil: Franco, Leonardo. Universidad de Málaga; España.
Fil: Juárez, Gustavo. Universidad Nacional de Tucumán; Argentina.
Fil: Gorrostieta, Efrén. Universidad Autónoma de Querétaro; México.
Fil: Orjuela-Cañon, Alvaro David. Universidad del Rosario; Colombia.
Sistemas de Automatización y Control
Materia
Bayesian approximation
Time series forecasting
Nonlinear autoregressive models
Recurrent neural networks
Nivel de accesibilidad
acceso abierto
Condiciones de uso
Repositorio
Repositorio Digital Universitario (UNC)
Institución
Universidad Nacional de Córdoba
OAI Identificador
oai:rdu.unc.edu.ar:11086/559207
Acceder
id RDUUNC_1d7bcbeb19223df7915226d1542ab56f
oai_identifier_str oai:rdu.unc.edu.ar:11086/559207
network_acronym_str RDUUNC
repository_id_str 2572
network_name_str Repositorio Digital Universitario (UNC)
spelling Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series ForecastingRodríguez Rivero, CristianPucheta, JuliánPatiño, DanielPuglisi, Jose LuisOtaño, PaulaFranco, LeonardoJuárez, GustavoGorrostieta, EfrénOrjuela-Cañon, Alvaro DavidBayesian approximationTime series forecastingNonlinear autoregressive modelsRecurrent neural networksFil: Rodríguez Rivero, Cristian. Universidad Nacional de Córdoba; Argentina.Fil: Pucheta, Julián. Universidad Nacional de Córdoba; Argentina.Fil: Patiño, Daniel. Universidad Nacional de San Juan. Instituto de Automática; Argentina.Fil: Puglisi, Jose Luis. Universidad de Carolina del Norte; Estados Unidos.Fil: Otaño, Paula. Universidad Tecnológica Nacional. Facultad Regional Córdoba.Fil: Franco, Leonardo. Universidad de Málaga; España.Fil: Juárez, Gustavo. Universidad Nacional de Tucumán; Argentina.Fil: Gorrostieta, Efrén. Universidad Autónoma de Querétaro; México.Fil: Orjuela-Cañon, Alvaro David. Universidad del Rosario; Colombia.For time series forecasting, obtaining models is based on the use of past observations from the same sequence. In those cases, when the model is learning from data, there is not an extra information that discuss about the quantity of noise inside the data available. In practice, it is necessary to deal with finite noisy datasets, which lead to uncertainty about the propriety of the model. For this problem, the employment of the Bayesian inference tools are preferable. A modified algorithm used for training a long-short term memory recurrent neural network for time series forecasting is presented. This approach was chosen to improve the forecasting of the original series, employing an implementation based on the minimization of the associated Kullback-Leibler Information Criterion. For comparison, a nonlinear autoregressive model implemented with a feedforward neural network was also presented. A simulation study was conducted to evaluate and illustrate results, comparing this approach with Bayesian neural-networks-based algorithms for artificial chaotic time-series and showing an improvement in terms of forecasting errors.https://www.springer.com/gp/book/9783030362102?wt_mc=ThirdParty.SpringerLink.3.EPR653.About_eBookpublishedVersionFil: Rodríguez Rivero, Cristian. Universidad Nacional de Córdoba; Argentina.Fil: Pucheta, Julián. Universidad Nacional de Córdoba; Argentina.Fil: Patiño, Daniel. Universidad Nacional de San Juan. Instituto de Automática; Argentina.Fil: Puglisi, Jose Luis. Universidad de Carolina del Norte; Estados Unidos.Fil: Otaño, Paula. Universidad Tecnológica Nacional. Facultad Regional Córdoba.Fil: Franco, Leonardo. Universidad de Málaga; España.Fil: Juárez, Gustavo. Universidad Nacional de Tucumán; Argentina.Fil: Gorrostieta, Efrén. Universidad Autónoma de Querétaro; México.Fil: Orjuela-Cañon, Alvaro David. Universidad del Rosario; Colombia.Sistemas de Automatización y Control2019info:eu-repo/semantics/bookPartinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_3248info:ar-repo/semantics/parteDeLibroapplication/pdf978-3-030-36211-9http://hdl.handle.net/11086/559207enginfo:eu-repo/semantics/openAccessreponame:Repositorio Digital Universitario (UNC)instname:Universidad Nacional de Córdobainstacron:UNC2025-12-04T09:22:30Zoai:rdu.unc.edu.ar:11086/559207Institucionalhttps://rdu.unc.edu.ar/Universidad públicaNo correspondehttp://rdu.unc.edu.ar/oai/snrdoca.unc@gmail.comArgentinaNo correspondeNo correspondeNo correspondeopendoar:25722025-12-04 09:22:30.784Repositorio Digital Universitario (UNC) - Universidad Nacional de Córdobafalse
dc.title.none.fl_str_mv Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting
title Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting
spellingShingle Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting
Rodríguez Rivero, Cristian
Bayesian approximation
Time series forecasting
Nonlinear autoregressive models
Recurrent neural networks
title_short Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting
title_full Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting
title_fullStr Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting
title_full_unstemmed Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting
title_sort Bayesian Inference for Training of Long Short Term Memory Models in Chaotic Time Series Forecasting
dc.creator.none.fl_str_mv Rodríguez Rivero, Cristian
Pucheta, Julián
Patiño, Daniel
Puglisi, Jose Luis
Otaño, Paula
Franco, Leonardo
Juárez, Gustavo
Gorrostieta, Efrén
Orjuela-Cañon, Alvaro David
author Rodríguez Rivero, Cristian
author_facet Rodríguez Rivero, Cristian
Pucheta, Julián
Patiño, Daniel
Puglisi, Jose Luis
Otaño, Paula
Franco, Leonardo
Juárez, Gustavo
Gorrostieta, Efrén
Orjuela-Cañon, Alvaro David
author_role author
author2 Pucheta, Julián
Patiño, Daniel
Puglisi, Jose Luis
Otaño, Paula
Franco, Leonardo
Juárez, Gustavo
Gorrostieta, Efrén
Orjuela-Cañon, Alvaro David
author2_role author
author
author
author
author
author
author
author
dc.subject.none.fl_str_mv Bayesian approximation
Time series forecasting
Nonlinear autoregressive models
Recurrent neural networks
topic Bayesian approximation
Time series forecasting
Nonlinear autoregressive models
Recurrent neural networks
dc.description.none.fl_txt_mv Fil: Rodríguez Rivero, Cristian. Universidad Nacional de Córdoba; Argentina.
Fil: Pucheta, Julián. Universidad Nacional de Córdoba; Argentina.
Fil: Patiño, Daniel. Universidad Nacional de San Juan. Instituto de Automática; Argentina.
Fil: Puglisi, Jose Luis. Universidad de Carolina del Norte; Estados Unidos.
Fil: Otaño, Paula. Universidad Tecnológica Nacional. Facultad Regional Córdoba.
Fil: Franco, Leonardo. Universidad de Málaga; España.
Fil: Juárez, Gustavo. Universidad Nacional de Tucumán; Argentina.
Fil: Gorrostieta, Efrén. Universidad Autónoma de Querétaro; México.
Fil: Orjuela-Cañon, Alvaro David. Universidad del Rosario; Colombia.
For time series forecasting, obtaining models is based on the use of past observations from the same sequence. In those cases, when the model is learning from data, there is not an extra information that discuss about the quantity of noise inside the data available. In practice, it is necessary to deal with finite noisy datasets, which lead to uncertainty about the propriety of the model. For this problem, the employment of the Bayesian inference tools are preferable. A modified algorithm used for training a long-short term memory recurrent neural network for time series forecasting is presented. This approach was chosen to improve the forecasting of the original series, employing an implementation based on the minimization of the associated Kullback-Leibler Information Criterion. For comparison, a nonlinear autoregressive model implemented with a feedforward neural network was also presented. A simulation study was conducted to evaluate and illustrate results, comparing this approach with Bayesian neural-networks-based algorithms for artificial chaotic time-series and showing an improvement in terms of forecasting errors.
https://www.springer.com/gp/book/9783030362102?wt_mc=ThirdParty.SpringerLink.3.EPR653.About_eBook
publishedVersion
Fil: Rodríguez Rivero, Cristian. Universidad Nacional de Córdoba; Argentina.
Fil: Pucheta, Julián. Universidad Nacional de Córdoba; Argentina.
Fil: Patiño, Daniel. Universidad Nacional de San Juan. Instituto de Automática; Argentina.
Fil: Puglisi, Jose Luis. Universidad de Carolina del Norte; Estados Unidos.
Fil: Otaño, Paula. Universidad Tecnológica Nacional. Facultad Regional Córdoba.
Fil: Franco, Leonardo. Universidad de Málaga; España.
Fil: Juárez, Gustavo. Universidad Nacional de Tucumán; Argentina.
Fil: Gorrostieta, Efrén. Universidad Autónoma de Querétaro; México.
Fil: Orjuela-Cañon, Alvaro David. Universidad del Rosario; Colombia.
Sistemas de Automatización y Control
description Fil: Rodríguez Rivero, Cristian. Universidad Nacional de Córdoba; Argentina.
publishDate 2019
dc.date.none.fl_str_mv 2019
dc.type.none.fl_str_mv info:eu-repo/semantics/bookPart
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_3248
info:ar-repo/semantics/parteDeLibro
format bookPart
status_str publishedVersion
dc.identifier.none.fl_str_mv 978-3-030-36211-9
http://hdl.handle.net/11086/559207
identifier_str_mv 978-3-030-36211-9
url http://hdl.handle.net/11086/559207
dc.language.none.fl_str_mv eng
language eng
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositorio Digital Universitario (UNC)
instname:Universidad Nacional de Córdoba
instacron:UNC
reponame_str Repositorio Digital Universitario (UNC)
collection Repositorio Digital Universitario (UNC)
instname_str Universidad Nacional de Córdoba
instacron_str UNC
institution UNC
repository.name.fl_str_mv Repositorio Digital Universitario (UNC) - Universidad Nacional de Córdoba
repository.mail.fl_str_mv oca.unc@gmail.com
_version_ 1850582855883685888
score 12.703443

Publicaciones similares