| On the convergence of multiplicative update algorithms for nonnegative matrix factorization. | |
| | |
MedLine Citation:
|
PMID: 18051179 Owner: NLM Status: In-Process |
Abstract/OtherAbstract:
|
Nonnegative matrix factorization (NMF) is useful to find basis information of nonnegative data. Currently, multiplicative updates are a simple and popular way to find the factorization. However, for the common NMF approach of minimizing the Euclidean distance between approximate and true values, no proof has shown that multiplicative updates converge to a stationary point of the NMF optimization problem. Stationarity is important as it is a necessary condition of a local minimum. This paper discusses the difficulty of proving the convergence. We propose slight modifications of existing updates and prove their convergence. Techniques invented in this paper may be applied to prove the convergence for other bound-constrained optimization problems. |
| | |
Authors:
|
Chih-Jen Lin |
Related Documents
:
|
2031649 - Implementing an aids policy. retaining and promoting productivity in the workplace. 21279299 - Skin problems of the stump in lower-limb amputees: 2. influence on functioning in daily... 12015879 - Finding motifs using random projections. 21566989 - The effect of vitamin-c infusion treatment on immune disorders - an invitation to a tri... 11829189 - Nonlinear digital hearing aid with near-instantaneous amplitude compression. |
Publication Detail:
|
Type: Journal Article; Research Support, Non-U.S. Gov't |
Journal Detail:
|
Title: IEEE transactions on neural networks / a publication of the IEEE Neural Networks Council Volume: 18 ISSN: 1045-9227 ISO Abbreviation: - Publication Date: 2007 Nov |
Date Detail:
|
Created Date: 2007-12-04 Completed Date: - Revised Date: - |
Medline Journal Info:
|
Nlm Unique ID: 101211035 Medline TA: IEEE Trans Neural Netw Country: United States |
Other Details:
|
Languages: eng Pagination: 1589-96 Citation Subset: IM |
Affiliation:
|
Department of Computer Science, National Taiwan University, Taipei 106, Taiwan. cjlin@csie.ntu.edu.tw |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
|
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Semiparametric regression using student t processes.
Next Document: An adaptable connectionist text-retrieval system with relevance feedback.