On restricted nonnegative matrix factorization
From MaRDI portal
Abstract: Nonnegative matrix factorization (NMF) is the problem of decomposing a given nonnegative matrix into a product of a nonnegative matrix and a nonnegative matrix . Restricted NMF requires in addition that the column spaces of and coincide. Finding the minimal inner dimension is known to be NP-hard, both for NMF and restricted NMF. We show that restricted NMF is closely related to a question about the nature of minimal probabilistic automata, posed by Paz in his seminal 1971 textbook. We use this connection to answer Paz's question negatively, thus falsifying a positive answer claimed in 1974. Furthermore, we investigate whether a rational matrix always has a restricted NMF of minimal inner dimension whose factors and are also rational. We show that this holds for matrices of rank at most and we exhibit a rank- matrix for which and require irrational entries.
Recommendations
Cited in
(12)- A geometric lower bound on the extension complexity of polytopes based on the f-vector
- Non-negative matrix factorization under equality constraints -- a study of industrial source identification
- scientific article; zbMATH DE number 5631077 (Why is no real title available?)
- On rationality of nonnegative matrix factorization
- Nonnegative matrix factorization requires irrationality
- On the complexity of recognizing nerves of convex sets
- The complexity of recognizing geometric hypergraphs
- On classifying continuous constraint satisfaction problems
- Representing matroids over the reals is \(\exists \mathbb{R}\)-complete
- Recognition of unit segment and polyline graphs is \(\exists \mathbb{R} \)-complete
- The complexity of recognizing geometric hypergraphs
- On reduced rank nonnegative matrix factorization for symmetric nonnegative matrices
This page was built for publication: On restricted nonnegative matrix factorization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4598244)