A convex model for non-negative matrix factorization and dimensionality reduction on physical space
Abstract
A collaborative convex framework for factoring a data matrix into a non-negative product , with a sparse coefficient matrix , is proposed. We restrict the columns of the dictionary matrix to coincide with certain columns of the data matrix , thereby guaranteeing a physically meaningful dictionary and dimensionality reduction. We use regularization to select the dictionary from the data and show this leads to an exact convex relaxation of in the case of distinct noise free data. We also show how to relax the restriction-to- constraint by initializing an alternating minimization approach with the solution of the convex model, obtaining a dictionary close to but not necessarily in . We focus on applications of the proposed framework to hyperspectral endmember and abundances identification and also show an application to blind source separation of NMR data.
Cite
@article{arxiv.1102.0844,
title = {A convex model for non-negative matrix factorization and dimensionality reduction on physical space},
author = {Ernie Esser and Michael Möller and Stanley Osher and Guillermo Sapiro and Jack Xin},
journal= {arXiv preprint arXiv:1102.0844},
year = {2015}
}
Comments
14 pages, 9 figures. EE and JX were supported by NSF grants {DMS-0911277}, {PRISM-0948247}, MM by the German Academic Exchange Service (DAAD), SO and MM by NSF grants {DMS-0835863}, {DMS-0914561}, {DMS-0914856} and ONR grant {N00014-08-1119}, and GS was supported by NSF, NGA, ONR, ARO, DARPA, and {NSSEFF.}