zbMATH — the first resource for mathematics

Projected gradient methods for nonnegative matrix factorization. (English) Zbl 1173.90583
Summary: Nonnegative matrix factorization (NMF) can be formulated as a minimization problem with bound constraints. Although bound-constrained optimization has been studied extensively in both theory and practice, so far no study has formally applied its techniques to NMF. In this letter, we propose two projected gradient methods for NMF, both of which exhibit strong optimization properties. We discuss efficient implementations and demonstrate that one of the proposed methods converges faster than the popular multiplicative update approach. A simple Matlab code is also provided.

90C52 Methods of reduced gradient type
90C30 Nonlinear programming
PDF BibTeX Cite
Full Text: DOI
[1] DOI: 10.1109/TAC.1976.1101194 · Zbl 0326.49025
[2] DOI: 10.1073/pnas.0308531101
[3] DOI: 10.1007/BF02592073 · Zbl 0634.90064
[4] DOI: 10.1016/S0167-6377(99)00074-7 · Zbl 0955.90128
[5] Hoyer P. O., Journal of Machine Learning Research 5 pp 1457– (2004)
[6] DOI: 10.1038/44565 · Zbl 1369.68285
[7] Lewis D. D., Journal of Machine Learning Research 5 pp 361– (2004)
[8] DOI: 10.1137/S1052623498345075 · Zbl 0957.65064
[9] DOI: 10.1137/0310009 · Zbl 0237.49019
[10] DOI: 10.2307/1390831
[11] DOI: 10.1002/env.3170050203
[12] DOI: 10.1007/BF01584660 · Zbl 0258.90043
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.