Cargando…

Bayesian optimization for computationally extensive probability distributions

An efficient method for finding a better maximizer of computationally extensive probability distributions is proposed on the basis of a Bayesian optimization technique. A key idea of the proposed method is to use extreme values of acquisition functions by Gaussian processes for the next training pha...

Descripción completa

Detalles Bibliográficos
Autores principales: Tamura, Ryo, Hukushima, Koji
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2018
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5837188/
https://www.ncbi.nlm.nih.gov/pubmed/29505596
http://dx.doi.org/10.1371/journal.pone.0193785
_version_ 1783304074541662208
author Tamura, Ryo
Hukushima, Koji
author_facet Tamura, Ryo
Hukushima, Koji
author_sort Tamura, Ryo
collection PubMed
description An efficient method for finding a better maximizer of computationally extensive probability distributions is proposed on the basis of a Bayesian optimization technique. A key idea of the proposed method is to use extreme values of acquisition functions by Gaussian processes for the next training phase, which should be located near a local maximum or a global maximum of the probability distribution. Our Bayesian optimization technique is applied to the posterior distribution in the effective physical model estimation, which is a computationally extensive probability distribution. Even when the number of sampling points on the posterior distributions is fixed to be small, the Bayesian optimization provides a better maximizer of the posterior distributions in comparison to those by the random search method, the steepest descent method, or the Monte Carlo method. Furthermore, the Bayesian optimization improves the results efficiently by combining the steepest descent method and thus it is a powerful tool to search for a better maximizer of computationally extensive probability distributions.
format Online
Article
Text
id pubmed-5837188
institution National Center for Biotechnology Information
language English
publishDate 2018
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-58371882018-03-19 Bayesian optimization for computationally extensive probability distributions Tamura, Ryo Hukushima, Koji PLoS One Research Article An efficient method for finding a better maximizer of computationally extensive probability distributions is proposed on the basis of a Bayesian optimization technique. A key idea of the proposed method is to use extreme values of acquisition functions by Gaussian processes for the next training phase, which should be located near a local maximum or a global maximum of the probability distribution. Our Bayesian optimization technique is applied to the posterior distribution in the effective physical model estimation, which is a computationally extensive probability distribution. Even when the number of sampling points on the posterior distributions is fixed to be small, the Bayesian optimization provides a better maximizer of the posterior distributions in comparison to those by the random search method, the steepest descent method, or the Monte Carlo method. Furthermore, the Bayesian optimization improves the results efficiently by combining the steepest descent method and thus it is a powerful tool to search for a better maximizer of computationally extensive probability distributions. Public Library of Science 2018-03-05 /pmc/articles/PMC5837188/ /pubmed/29505596 http://dx.doi.org/10.1371/journal.pone.0193785 Text en © 2018 Tamura, Hukushima http://creativecommons.org/licenses/by/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
spellingShingle Research Article
Tamura, Ryo
Hukushima, Koji
Bayesian optimization for computationally extensive probability distributions
title Bayesian optimization for computationally extensive probability distributions
title_full Bayesian optimization for computationally extensive probability distributions
title_fullStr Bayesian optimization for computationally extensive probability distributions
title_full_unstemmed Bayesian optimization for computationally extensive probability distributions
title_short Bayesian optimization for computationally extensive probability distributions
title_sort bayesian optimization for computationally extensive probability distributions
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5837188/
https://www.ncbi.nlm.nih.gov/pubmed/29505596
http://dx.doi.org/10.1371/journal.pone.0193785
work_keys_str_mv AT tamuraryo bayesianoptimizationforcomputationallyextensiveprobabilitydistributions
AT hukushimakoji bayesianoptimizationforcomputationallyextensiveprobabilitydistributions