Comparison between stochastic gradient descent and VLE metaheuristic for optimizing matrix factorization

Juan A. Gómez-Pulido, Enrique Cortés-Toro, Arturo Durán-Domínguez, José M. Lanza-Gutiérrez, BRODERICK CRAWFORD LABRIN, RICARDO JAVIER SOTO DE GIORGIS

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Matrix factorization is used by recommender systems in collaborative filtering for building prediction models based on a couple of matrices. These models are usually generated by stochastic gradient descent algorithm, which learns the model minimizing the error done. Finally, the obtained models are validated according to an error criterion by predicting test data. Since the model generation can be tackled as an optimization problem where there is a huge set of possible solutions, we propose to use metaheuristics as alternative solving methods for matrix factorization. In this work we applied a novel metaheuristic for continuous optimization, which works inspired by the vapour-liquid equilibrium. We considered a particular case were matrix factorization was applied: the prediction student performance problem. The obtained results surpassed thoroughly the accuracy provided by stochastic gradient descent.

Original languageEnglish
Title of host publicationOptimization and Learning - 3rd International Conference, OLA 2020, Proceedings
EditorsBernabé Dorronsoro, Patricia Ruiz, Juan Carlos de la Torre, Daniel Urda, El-Ghazali Talbi
PublisherSpringer
Pages153-164
Number of pages12
ISBN (Print)9783030419127
DOIs
StatePublished - 2020
Externally publishedYes
Event3rd International Conference on Optimization and Learning, OLA 2020 - Cádiz, Spain
Duration: 17 Feb 202019 Feb 2020

Publication series

NameCommunications in Computer and Information Science
Volume1173 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference3rd International Conference on Optimization and Learning, OLA 2020
CountrySpain
CityCádiz
Period17/02/2019/02/20

Keywords

  • Gradient descent
  • Matrix factorization
  • Metaheuristics

Fingerprint Dive into the research topics of 'Comparison between stochastic gradient descent and VLE metaheuristic for optimizing matrix factorization'. Together they form a unique fingerprint.

Cite this