SVD based initialization: A head start for nonnegative matrix factorization

PATTERN RECOGNITION

卷:41

Article

Boutsidis, C.¹  Gallopoulos, E.² 

[1] Rensselaer Polytech Inst, Dept Comp Sci, Troy, NY 12180 USA

[2] Univ Patras, Comp Engn & Informat Dept, GR-26500 Patras, Greece

关键词: NMF;    sparse NMF;    SVD;    nonnegative matrix factorization;    singular value decomposition;    Perron-Frobenius;    low rank;    structured initialization;    sparse factorization;   

DOI  :  10.1016/j.patcog.2007.09.010

来源: Elsevier

【 摘 要 】

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. (c) 2007 Elsevier Ltd. All rights reserved.

【 授权许可】

Free