The NMF problem and lattice-subspaces
The nonnegative matrix factorization (NMF) problem for a nonnegative \(n\times m\) real matrix \(A\) is the problem of determining two nonnegative real \(n\times p\) matrix \(F\) and \(p\times m\) matrix \(V\) such that \(A =FV\) with \(p<\min \{n, m\}\). When \(p\) is the rank of \(A\), the problem is known as nonnegative rank factorization (NRF) of \(A\) and some results have existed in the literature. The problem does not have always a solution. In this paper, an algorithmic method is given for finding \(F, V\) such that the ranks of \(F\) and \(A\) are the same, under some condition on \(p\). The method is based on the theory of lattice-subspaces and positive bases, namely, it involves a positive basis of a minimal lattice-subspace which contains a maximal set of linearly independent rows of \(A\). The author notes that when \(A\) is symmetric and nonnegative, the method does not warrant a symmetric factorization \(A = WW^T\). Examples are provided to illustrate the algorithm which has different cases. A MATLAB code is used in some of the examples. An appendix is given on lattice-subspaces and positive bases.
- Nonnegative rank factorization -- a heuristic approach via rank reduction
- On reduced rank nonnegative matrix factorization for symmetric nonnegative matrices
- Computing symmetric nonnegative rank factorizations
- On the set of solutions of the nonnegative matrix factorization problem
- Nonnegative matrix factorization
- A reverse Hadamard inequality
- Algorithms and applications for approximate nonnegative matrix factorization
- Computation of vector sublattices and minimal lattice-subspaces of \(\mathbb R^k\): applications in finance
- Computing nonnegative rank factorizations
- Computing symmetric nonnegative rank factorizations
- Finite-dimensional lattice-subspaces of 𝐶(Ω) and curves of ℝⁿ
- Heuristics for exact nonnegative matrix factorization
- Minimal lattice-subspaces
- Nonnegative Matrix Factorization Based on Alternating Nonnegativity Constrained Least Squares and Active Set Method
- Nonnegative ranks, decompositions, and factorizations of nonnegative matrices
- On nonnegative factorization of matrices
- On the geometric interpretation of the nonnegative rank
- SVD based initialization: A head start for nonnegative matrix factorization
- The nonnegative rank of a matrix: hard problems, easy solutions
- The quickhull algorithm for convex hulls
- Positive matrix factorization via extremal polyhedral cones
- Nonnegative rank factorization -- a heuristic approach via rank reduction
- On the set of solutions of the nonnegative matrix factorization problem
- scientific article; zbMATH DE number 802815 (Why is no real title available?)
- Lattice Identification and Separation: Theory and Algorithm
- Nonnegative matrix factorization requires irrationality
- Computing symmetric nonnegative rank factorizations
This page was built for publication: The NMF problem and lattice-subspaces
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2185834)
