Cargando…

Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation †

We consider estimating the marginal likelihood in settings with independent and identically distributed (i.i.d.) data. We propose estimating the predictive distributions in a sequential factorization of the marginal likelihood in such settings by using stochastic gradient Markov Chain Monte Carlo te...

Descripción completa

Detalles Bibliográficos
Autores principales: Cameron, Scott A., Eggers, Hans C., Kroon, Steve
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7514453/
http://dx.doi.org/10.3390/e21111109
_version_ 1783586591666601984
author Cameron, Scott A.
Eggers, Hans C.
Kroon, Steve
author_facet Cameron, Scott A.
Eggers, Hans C.
Kroon, Steve
author_sort Cameron, Scott A.
collection PubMed
description We consider estimating the marginal likelihood in settings with independent and identically distributed (i.i.d.) data. We propose estimating the predictive distributions in a sequential factorization of the marginal likelihood in such settings by using stochastic gradient Markov Chain Monte Carlo techniques. This approach is far more efficient than traditional marginal likelihood estimation techniques such as nested sampling and annealed importance sampling due to its use of mini-batches to approximate the likelihood. Stability of the estimates is provided by an adaptive annealing schedule. The resulting stochastic gradient annealed importance sampling (SGAIS) technique, which is the key contribution of our paper, enables us to estimate the marginal likelihood of a number of models considerably faster than traditional approaches, with no noticeable loss of accuracy. An important benefit of our approach is that the marginal likelihood is calculated in an online fashion as data becomes available, allowing the estimates to be used for applications such as online weighted model combination.
format Online
Article
Text
id pubmed-7514453
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-75144532020-11-09 Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation † Cameron, Scott A. Eggers, Hans C. Kroon, Steve Entropy (Basel) Article We consider estimating the marginal likelihood in settings with independent and identically distributed (i.i.d.) data. We propose estimating the predictive distributions in a sequential factorization of the marginal likelihood in such settings by using stochastic gradient Markov Chain Monte Carlo techniques. This approach is far more efficient than traditional marginal likelihood estimation techniques such as nested sampling and annealed importance sampling due to its use of mini-batches to approximate the likelihood. Stability of the estimates is provided by an adaptive annealing schedule. The resulting stochastic gradient annealed importance sampling (SGAIS) technique, which is the key contribution of our paper, enables us to estimate the marginal likelihood of a number of models considerably faster than traditional approaches, with no noticeable loss of accuracy. An important benefit of our approach is that the marginal likelihood is calculated in an online fashion as data becomes available, allowing the estimates to be used for applications such as online weighted model combination. MDPI 2019-11-12 /pmc/articles/PMC7514453/ http://dx.doi.org/10.3390/e21111109 Text en © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Cameron, Scott A.
Eggers, Hans C.
Kroon, Steve
Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation †
title Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation †
title_full Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation †
title_fullStr Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation †
title_full_unstemmed Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation †
title_short Stochastic Gradient Annealed Importance Sampling for Efficient Online Marginal Likelihood Estimation †
title_sort stochastic gradient annealed importance sampling for efficient online marginal likelihood estimation †
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7514453/
http://dx.doi.org/10.3390/e21111109
work_keys_str_mv AT cameronscotta stochasticgradientannealedimportancesamplingforefficientonlinemarginallikelihoodestimation
AT eggershansc stochasticgradientannealedimportancesamplingforefficientonlinemarginallikelihoodestimation
AT kroonsteve stochasticgradientannealedimportancesamplingforefficientonlinemarginallikelihoodestimation