Abstract | ||
---|---|---|
We describe Nonnegative Double Singular Value Decomposition (NNDSVD), a new method designed to enhance the initialization stage of nonnegative matrix factorization (NMF). NNDSVD can readily be combined with existing NMF algorithms. The basic algorithm contains no randomization and is based on two SVD processes, one approximating the data matrix, the other approximating positive sections of the resulting partial SVD factors utilizing an algebraic property of unit rank matrices. Simple practical variants for NMF with dense factors are described. NNDSVD is also well suited to initialize NMF algorithms with sparse factors. Many numerical examples suggest that NNDSVD leads to rapid reduction of the approximation error of many NMF algorithms. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1016/j.patcog.2007.09.010 | Pattern Recognition |
Keywords | Field | DocType |
nmf,svd,singular value decomposition,approximating positive section,svd process,nmf algorithm,structured initialization,low rank,head start,unit rank matrix,perron-frobenius,partial svd factor,algebraic property,approximation error,nonnegative matrix factorization,sparse factorization,sparse nmf,data matrix | Algebraic number,Computer simulation,Matrix (mathematics),Artificial intelligence,Singular value decomposition,Mathematical optimization,Pattern recognition,Matrix decomposition,Algorithm,Non-negative matrix factorization,Initialization,Approximation error,Mathematics | Journal |
Volume | Issue | ISSN |
41 | 4 | Pattern Recognition |
Citations | PageRank | References |
149 | 6.36 | 21 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Christos Boutsidis | 1 | 610 | 33.37 |
E. Gallopoulos | 2 | 202 | 19.79 |