Perturbation of matrices and non-negative rank with a view toward statistical models

Cristiano Bocci, Enrico Carlini, Fabio Rapallo

Published: 2010-10-18 | Updated: 2011-07-20

Comments: 13 pages, 3 figures. A theorem has been rewritten, and some improvements in the presentations have been implemented

Note:

DOI: 10.1137/110825455

statistical model

# BIB PDF ARXIV

Abstract

In this paper we study how perturbing a matrix changes its non-negative rank. We prove that the non-negative rank is upper-semicontinuos and we describe some special families of perturbations. We show how our results relate to Statistics in terms of the study of Maximum Likelihood Estimation for mixture models.

arXiv++ Combinatorics

Perturbation of matrices and non-negative rank with a view toward statistical models

Abstract

BibTeX