Semidefinite Programming Based Preconditioning for More Robust Near-Separable Nonnegative Matrix Factorization
Preprint
Abstract
Nonnegative matrix factorization (NMF) under the separability assumption can provably be solved efficiently, even in the presence of noise, and has been shown to be a powerful technique in document classification and hyperspectral unmixing. This problem is referred to as near-separable NMF and requires that there exists a cone spanned by a small subset of the columns of the input nonnegative matrix approximately containing all columns. In this paper, we propose a preconditioning based on semidefinite programming making the input matrix well-conditioned. This in turn can improve significantly the performance of near-separable NMF algorithms which is illustrated on the popular successive projection algorithm (SPA). The new preconditioned SPA is provably more robust to noise, and outperforms SPA on several synthetic data sets. We also show how an active-set method allow us to apply the preconditioning on large-scale real-world hyperspectral images.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- A Signal Processing Perspective on Hyperspectral Unmixing: Insights from Remote SensingIEEE Signal Processing Magazine, 2013
- Linear convergence of a modified Frank–Wolfe algorithm for computing minimum-volume enclosing ellipsoidsOptimization Methods and Software, 2008
- On Khachiyan's algorithm for the computation of minimum-volume enclosing ellipsoidsDiscrete Applied Mathematics, 2007
- Nonnegative matrix factorization for spectral data analysisLinear Algebra and its Applications, 2006
- Minimum-Volume Enclosing Ellipsoids and Core SetsJournal of Optimization Theory and Applications, 2005
- Vertex component analysis: a fast algorithm to unmix hyperspectral dataIEEE Transactions on Geoscience and Remote Sensing, 2005
- The successive projections algorithm for variable selection in spectroscopic multicomponent analysisChemometrics and Intelligent Laboratory Systems, 2001
- Learning the parts of objects by non-negative matrix factorizationNature, 1999
- On the complexity of approximating the maximal inscribed ellipsoid for a polytopeMathematical Programming, 1993
- Generalized equations and their solutions, part II: Applications to nonlinear programming1982