Development of a hyperparameter optimization method for recommendatory models based on matrix factorization

Many advanced recommendatory models are implemented using matrix factorization algorithms. Experiments show that the quality of their performance depends significantly on the selected hyperparameters. Analysis of the effectiveness of using various methods for solving this problem of optimizing hyper...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Alexander Nechaev, Vasily Meltsov, Dmitry Strabykin
Formato:	article
Lenguaje:	EN RU UK
Publicado:	PC Technology Center 2021
Materias:	bayesian optimization gaussian process covariance function matrix factorization recommendatory systems Technology (General) T1-995 Industry HD2321-4730.9
Acceso en línea:	https://doaj.org/article/deec9da69c3948b888fcd52ab029a4b2
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:deec9da69c3948b888fcd52ab029a4b2
record_format	dspace
spelling	oai:doaj.org-article:deec9da69c3948b888fcd52ab029a4b22021-11-04T14:06:46ZDevelopment of a hyperparameter optimization method for recommendatory models based on matrix factorization1729-37741729-406110.15587/1729-4061.2021.239124https://doaj.org/article/deec9da69c3948b888fcd52ab029a4b22021-10-01T00:00:00Zhttp://journals.uran.ua/eejet/article/view/239124https://doaj.org/toc/1729-3774https://doaj.org/toc/1729-4061Many advanced recommendatory models are implemented using matrix factorization algorithms. Experiments show that the quality of their performance depends significantly on the selected hyperparameters. Analysis of the effectiveness of using various methods for solving this problem of optimizing hyperparameters was made. It has shown that the use of classical Bayesian optimization which treats the model as a «black box» remains the standard solution. However, the models based on matrix factorization have a number of characteristic features. Their use makes it possible to introduce changes in the optimization process leading to a decrease in the time required to find the sought points without losing quality. Modification of the Gaussian process core which is used as a surrogate model for the loss function when performing the Bayesian optimization was proposed. The described modification at first iterations increases the variance of the values predicted by the Gaussian process over a given region of the hyperparameter space. In some cases, this makes it possible to obtain more information about the real form of the investigated loss function in less time. Experiments were carried out using well-known data sets for recommendatory systems. Total optimization time when applying the modification was reduced by 16 % (or 263 seconds) at best and remained the same at worst (less than 1-second difference). In this case, the expected error of the recommendatory model did not change (the absolute difference in values is two orders of magnitude lower than the value of error reduction in the optimization process). Thus, the use of the proposed modification contributes to finding a better set of hyperparameters in less time without loss of qualityAlexander NechaevVasily MeltsovDmitry StrabykinPC Technology Centerarticlebayesian optimizationgaussian processcovariance functionmatrix factorizationrecommendatory systemsTechnology (General)T1-995IndustryHD2321-4730.9ENRUUKEastern-European Journal of Enterprise Technologies, Vol 5, Iss 4 (113), Pp 45-54 (2021)
institution	DOAJ
collection	DOAJ
language	EN RU UK
topic	bayesian optimization gaussian process covariance function matrix factorization recommendatory systems Technology (General) T1-995 Industry HD2321-4730.9
spellingShingle	bayesian optimization gaussian process covariance function matrix factorization recommendatory systems Technology (General) T1-995 Industry HD2321-4730.9 Alexander Nechaev Vasily Meltsov Dmitry Strabykin Development of a hyperparameter optimization method for recommendatory models based on matrix factorization
description	Many advanced recommendatory models are implemented using matrix factorization algorithms. Experiments show that the quality of their performance depends significantly on the selected hyperparameters. Analysis of the effectiveness of using various methods for solving this problem of optimizing hyperparameters was made. It has shown that the use of classical Bayesian optimization which treats the model as a «black box» remains the standard solution. However, the models based on matrix factorization have a number of characteristic features. Their use makes it possible to introduce changes in the optimization process leading to a decrease in the time required to find the sought points without losing quality. Modification of the Gaussian process core which is used as a surrogate model for the loss function when performing the Bayesian optimization was proposed. The described modification at first iterations increases the variance of the values predicted by the Gaussian process over a given region of the hyperparameter space. In some cases, this makes it possible to obtain more information about the real form of the investigated loss function in less time. Experiments were carried out using well-known data sets for recommendatory systems. Total optimization time when applying the modification was reduced by 16 % (or 263 seconds) at best and remained the same at worst (less than 1-second difference). In this case, the expected error of the recommendatory model did not change (the absolute difference in values is two orders of magnitude lower than the value of error reduction in the optimization process). Thus, the use of the proposed modification contributes to finding a better set of hyperparameters in less time without loss of quality
format	article
author	Alexander Nechaev Vasily Meltsov Dmitry Strabykin
author_facet	Alexander Nechaev Vasily Meltsov Dmitry Strabykin
author_sort	Alexander Nechaev
title	Development of a hyperparameter optimization method for recommendatory models based on matrix factorization
title_short	Development of a hyperparameter optimization method for recommendatory models based on matrix factorization
title_full	Development of a hyperparameter optimization method for recommendatory models based on matrix factorization
title_fullStr	Development of a hyperparameter optimization method for recommendatory models based on matrix factorization
title_full_unstemmed	Development of a hyperparameter optimization method for recommendatory models based on matrix factorization
title_sort	development of a hyperparameter optimization method for recommendatory models based on matrix factorization
publisher	PC Technology Center
publishDate	2021
url	https://doaj.org/article/deec9da69c3948b888fcd52ab029a4b2
work_keys_str_mv	AT alexandernechaev developmentofahyperparameteroptimizationmethodforrecommendatorymodelsbasedonmatrixfactorization AT vasilymeltsov developmentofahyperparameteroptimizationmethodforrecommendatorymodelsbasedonmatrixfactorization AT dmitrystrabykin developmentofahyperparameteroptimizationmethodforrecommendatorymodelsbasedonmatrixfactorization
_version_	1718444872577318912

Development of a hyperparameter optimization method for recommendatory models based on matrix factorization

Ejemplares similares