Compressed Nonnegative Matrix Factorization Is Fast and Accurate

DOI10.1109/TSP.2016.2516971MaRDI QIDQ4618319zbMATH OpenFDO

Authors Mariano Tepper, Guillermo Sapiro

Publication date 7 February 2019

Published in IEEE Transactions on Signal Processing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1505.04650

Mathematics Subject Classification ID

Factorization of matrices (15A23) Positive matrices and their generalizations; cones of matrices (15B48)

Abstract: Nonnegative matrix factorization (NMF) has an established reputation as a useful data analysis technique in numerous applications. However, its usage in practical situations is undergoing challenges in recent years. The fundamental factor to this is the increasingly growing size of the datasets available and needed in the information sciences. To address this, in this work we propose to use structured random compression, that is, random projections that exploit the data structure, for two NMF variants: classical and separable. In separable NMF (SNMF) the left factors are a subset of the columns of the input matrix. We present suitable formulations for each problem, dealing with different representative algorithms within each one. We show that the resulting compressed techniques are faster than their uncompressed variants, vastly reduce memory demands, and do not encompass any significant deterioration in performance. The proposed structured random projections for SNMF allow to deal with arbitrarily shaped large matrices, beyond the standard limit of tall-and-skinny matrices, granting access to very efficient computations in this general setting. We accompany the algorithmic presentation with theoretical foundations and numerous and diverse examples, showing the suitability of the proposed approaches.

Cited in

(6)

• Efficient preconditioning for noisy separable nonnegative matrix factorization problems by successive projection based low-rank approximations
• Clustering is semidefinitely not that hard: nonnegative SDP for manifold disentangling
• A goodness-of-fit test on the number of biclusters in a relational data matrix
• Randomized algorithms for orthogonal nonnegative matrix factorization
• Randomized low-rank approximation methods for projection-based model order reduction of large nonlinear dynamical problems
• Randomized algorithms for symmetric nonnegative matrix factorization

This page was built for publication: Compressed Nonnegative Matrix Factorization Is Fast and Accurate

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4618319)