Positive tensor factorization

被引：123

作者：

Welling, M ^{[1
]}

Weber, M ^{[1
]}

机构：

[1] CALTECH, Pasadena, CA 91125 USA

来源：

PATTERN RECOGNITION LETTERS | 2001年 / 22卷 / 12期

关键词：

PCA; SVD; positive matrix factorization; feature extraction;

D O I：

10.1016/S0167-8655(01)00070-8

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A novel fixed point algorithm for positive tensor factorization (PTF) is introduced. The update rules efficiently minimize the reconstruction error of a positive tensor over positive factors. Tensors of arbitrary order can be factorized, which extends earlier results in the literature. Experiments show that the factors of PTF are easier to interpret than those produced by methods based on the singular value decomposition, which might contain negative values. We also illustrate the tendency of PTF to generate sparsely distributed codes. (C) 2001 Published by Elsevier Science B.V.

引用

页码：1255 / 1261

页数：7

共 7 条

[1] RELATIONS BETWEEN THE STATISTICS OF NATURAL IMAGES AND THE RESPONSE PROPERTIES OF CORTICAL-CELLS
FIELD, DJ
[J]. JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION, 1987, 4 (12): : 2379 - 2394
[2] Generative models for discovering sparse distributed representations
Hinton, GE
Ghahramani, Z
[J]. PHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY OF LONDON SERIES B-BIOLOGICAL SCIENCES, 1997, 352 (1358) : 1177 - 1190
[3] Learning the parts of objects by non-negative matrix factorization
Lee, DD
Seung, HS
[J]. NATURE, 1999, 401 (6755) : 788 - 791
[4] Least squares formulation of robust non-negative factor analysis
Paatero, P
[J]. CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 1997, 37 (01) : 23 - 35
[5] PERONA P, 1998, P EUR C COMP VIS, P655
[6] ROWEIS S, 1997, NEURAL INFORM PROCES, V10, P626
[7] SHY D, 1994, 1994 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, P237, DOI 10.1109/CVPR.1994.323835

← 1 →