Matrix Factorization based Collaborative Filtering with Resilient Stochastic Gradient Descent
Abstract
One of the leading approaches to collaborative filtering is to use matrix factorization to discover a set of latent factors that explain the pattern of preferences. In this paper, we apply a resilient stochastic gradient descent approach that uses only the sign of the gradient, similar to the R-Prop algorithm in neural network training, to matrix factorization for collaborative filtering. We evaluate the performance of our approach on the MovieLens 1M dataset, and find that test set accuracy markedly improves compared to standard gradient descent. As a follow-up experiment, we apply clustering to the learned item-factor matrix in factor space, and attempt to manually characterize each cluster of movies.
Recommended Citation
A. M. Abdelbar et al., "Matrix Factorization based Collaborative Filtering with Resilient Stochastic Gradient Descent," Proceedings of the International Joint Conference on Neural Networks (2018, Rio de Janeiro, Brazil), Institute of Electrical and Electronics Engineers (IEEE), Jul 2018.
The definitive version is available at https://doi.org/10.1109/IJCNN.2018.8489528
Meeting Name
2018 International Joint Conference on Neural Networks, IJCNN 2018 (2018: Jul. 8-13, Rio de Janeiro, Brazil)
Department(s)
Electrical and Computer Engineering
Research Center/Lab(s)
Center for High Performance Computing Research
Keywords and Phrases
Factorization; Matrix Algebra; Neural Networks; Statistical Tests; Stochastic Systems, Factor Space; Gradient Descent; Latent Factor; Matrix Factorizations; Movielens; Neural Network Training; Stochastic Gradient Descent; Test Sets, Collaborative Filtering
International Standard Book Number (ISBN)
978-1-5090-6014-6
International Standard Serial Number (ISSN)
2161-4407
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2018 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.
Publication Date
01 Jul 2018
Comments
Partial support from the Brandon University Research Council (BURC), the Missouri University of Science and Technology Intelligent Systems Center and the M.K. Finley Missouri Endowment is gratefully acknowledged.