English

Nonnegative Matrix Factorization Requires Irrationality

Computational Complexity 2017-03-24 v2 Machine Learning Numerical Analysis

Abstract

Nonnegative matrix factorization (NMF) is the problem of decomposing a given nonnegative n×mn \times m matrix MM into a product of a nonnegative n×dn \times d matrix WW and a nonnegative d×md \times m matrix HH. A longstanding open question, posed by Cohen and Rothblum in 1993, is whether a rational matrix MM always has an NMF of minimal inner dimension dd whose factors WW and HH are also rational. We answer this question negatively, by exhibiting a matrix for which WW and HH require irrational entries.

Keywords

Cite

@article{arxiv.1605.06848,
  title  = {Nonnegative Matrix Factorization Requires Irrationality},
  author = {Dmitry Chistikov and Stefan Kiefer and Ines Marušić and Mahsa Shirmohammadi and James Worrell},
  journal= {arXiv preprint arXiv:1605.06848},
  year   = {2017}
}

Comments

Journal version, to appear in the SIAM Journal on Applied Algebra and Geometry (SIAGA)

R2 v1 2026-06-22T14:06:49.278Z