Robust near-separable nonnegative matrix factorization using linear optimization

MaRDI QIDQ2934034zbMATH OpenFDO

Authors Nicolas Gillis, Robert Luce

Publication date 8 December 2014

Full work available at URL https://arxiv.org/abs/1302.4385, http://jmlr.csail.mit.edu/papers/v15/gillis14a.html

zbMATH Keywords

matrix factorizationlinear programmingconvex optimizationhyperspectral unmixingrobustness to noiseseparabilitypure-pixel assumption

Mathematics Subject Classification ID

Multivariate analysis (62H99) Linear programming (90C05) Factorization of matrices (15A23)

Abstract: Nonnegative matrix factorization (NMF) has been shown recently to be tractable under the separability assumption, under which all the columns of the input data matrix belong to the convex cone generated by only a few of these columns. Bittorf, Recht, R'e and Tropp (`Factoring nonnegative matrices with linear programs', NIPS 2012) proposed a linear programming (LP) model, referred to as Hottopixx, which is robust under any small perturbation of the input matrix. However, Hottopixx has two important drawbacks: (i) the input matrix has to be normalized, and (ii) the factorization rank has to be known in advance. In this paper, we generalize Hottopixx in order to resolve these two drawbacks, that is, we propose a new LP model which does not require normalization and detects the factorization rank automatically. Moreover, the new LP model is more flexible, significantly more tolerant to noise, and can easily be adapted to handle outliers and other noise models. Finally, we show on several synthetic datasets that it outperforms Hottopixx while competing favorably with two state-of-the-art methods.

Recommendations

• Robustness analysis of Hottopixx, a linear programming model for factoring nonnegative matrices
• Refinement of Hottopixx method for nonnegative matrix factorization under noisy separability
• Nonnegative matrix factorization via archetypal analysis
• Robustness analysis of preconditioned successive projection algorithm for general form of separable NMF problem
• Semidefinite programming based preconditioning for more robust near-separable nonnegative matrix factorization

Cited in

(14)

• Simplex-Structured Matrix Factorization: Sparsity-Based Identifiability and Provably Correct Algorithms
• Coseparable Nonnegative Matrix Factorization
• Nonnegative matrix factorization via archetypal analysis
• Ellipsoidal rounding for nonnegative matrix factorization under noisy separability
• Enhancing pure-pixel identification performance via preconditioning
• Semidefinite programming based preconditioning for more robust near-separable nonnegative matrix factorization
• Smoothed separable nonnegative matrix factorization
• Refinement of Hottopixx method for nonnegative matrix factorization under noisy separability
• Convex nonnegative matrix factorization with manifold regularization
• An alternating direction and projection algorithm for structure-enforced matrix factorization
• Provably robust blind source separation of linear-quadratic near-separable mixtures
• Analyzing Raman spectral data without separabiliy assumption
• Robustness analysis of Hottopixx, a linear programming model for factoring nonnegative matrices
• On the robustness of the successive projection algorithm

This page was built for publication: Robust near-separable nonnegative matrix factorization using linear optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2934034)