An extension of neural gas to local PCA

被引：31

作者：

Möller, R

Hoffmann, H

机构：

[1] Univ Bielefeld, Fac Technol, Comp Engn Grp, D-33594 Bielefeld, Germany

[2] Max Planck Inst Psychol Res, D-80799 Munich, Germany

来源：

NEUROCOMPUTING | 2004年 / 62卷

关键词：

unsupervised learning; local principal component analysis; vector quantization; neural gas; handwritten digit recognition;

D O I：

10.1016/j.neucom.2003.09.014

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We suggest an extension of the neural gas vector quantization method to local principal component analysis. The distance measure for the competition between local units combines a normalized Mahalanobis distance in the principal subspace and the squared reconstruction error, with the weighting of both measures depending on the residual variance in the minor subspace. A recursive least-squares method performs the local principal component analysis. The method is tested on synthetic two- and three-dimensional data and on the recognition of handwritten digits. (C) 2004 Elsevier B.V. All rights reserved.

引用

页码：305 / 326

页数：22

共 27 条

[1] Generalized radial basis function networks for classification and novelty detection: self-organization of optimal Bayesian decision
Albrecht, S
Busch, J
Kloppenburg, M
Metze, F
Tavan, P
[J]. NEURAL NETWORKS, 2000, 13 (10) : 1075 - 1093
[2] Soft vector quantization and the EM algorithm
Alpaydin, E
[J]. NEURAL NETWORKS, 1998, 11 (03) : 467 - 477
[3] PRINCIPAL COMPONENT EXTRACTION USING RECURSIVE LEAST-SQUARES LEARNING
BANNOUR, S
AZIMISADJADI, MR
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1995, 6 (02): : 457 - 469
[4] Bregler C., 1995, Advances in Neural Information Processing Systems 7, P973
[5] MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM
DEMPSTER, AP
LAIRD, NM
RUBIN, DB
[J]. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01): : 1 - 38
[6] Diamantaras KI, 1996, Principal Component Neural Networks: Theory and Applications
[7] Image segmentation using a mixture of principal components representation
Dony, RD
Haykin, S
[J]. IEE PROCEEDINGS-VISION IMAGE AND SIGNAL PROCESSING, 1997, 144 (02): : 73 - 80
[8] Golub G.H., 2013, Matrix Computations, V4th
[9] THE PRINCIPAL COMPONENTS OF NATURAL IMAGES
HANCOCK, PJB
BADDELEY, RJ
SMITH, LS
[J]. NETWORK-COMPUTATION IN NEURAL SYSTEMS, 1992, 3 (01) : 61 - 70
[10] Modeling the manifolds of images of handwritten digits
Hinton, GE
Dayan, P
Revow, M
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1997, 8 (01): : 65 - 74

← 1 2 3 →