Automatic Hyperparameter Tuning in Sparse Matrix Factorization

DOI10.1162/NECO_A_01581MaRDI QIDQ6136208zbMATH OpenFDO

Authors Ryota Kawasumi, Koujin Takeda

Publication date 28 August 2023

Published in Neural Computation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2305.10114

zbMATH Keywords

Bayesian frameworksparse matrix factorizationhyperparameter tuning

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Bayesian inference (62F15) Factor analysis and principal components; correspondence analysis (62H25) Computational methods for sparse matrices (65F50) Factorization of matrices (15A23) Numerical computation of eigenvalues and eigenvectors of matrices (65F15)

Abstract: We study the problem of hyperparameter tuning in sparse matrix factorization under Bayesian framework. In the prior work, an analytical solution of sparse matrix factorization with Laplace prior was obtained by variational Bayes method under several approximations. Based on this solution, we propose a novel numerical method of hyperparameter tuning by evaluating the zero point of normalization factor in sparse matrix prior. We also verify that our method shows excellent performance for ground-truth sparse matrix reconstruction by comparing it with the widely-used algorithm of sparse principal component analysis.

Recommendations

• Approximate method of variational Bayesian matrix factorization/completion with sparse prior
• Adaptive sparse regression
• Online learning for matrix factorization and sparse coding
• Variational Bayesian sparse additive matrix factorization
• Least squares auto-tuning

This page was built for publication: Automatic Hyperparameter Tuning in Sparse Matrix Factorization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6136208)