Topological local principal component analysis

Author

Liu, Zhi-Yong ; Xu, Lei

Author_Institution

Dept. of Comput. Sci. & Eng., Chinese Univ. of Hong Kong, China

Volume

3

fYear

2002

fDate

18-22 Nov. 2002

Firstpage

1346

Abstract

We propose a topological local principal component analysis (PCA) in help of Kohonen´s self-organizing maps (SOM). The topological local PCA describes one cluster by one neuron such that it is capable of exploiting both the global topological structure and each local cluster structure. We also investigate a newly proposed self-organizing strategy that can enhance the learning speed, as well as an alternative Stiefel manifold based algorithm to ensure the orthonormality constraint of the local PCA.

Keywords

Gaussian distribution; principal component analysis; self-organising feature maps; topology; unsupervised learning; Gaussian mixture model; Kohonen self-organizing maps; Stiefel manifold; global topological structure; learning speed; local cluster structure; orthonormality constraint; topological local principal component analysis; Clustering algorithms; Computer science; Councils; Covariance matrix; Feature extraction; Iterative algorithms; Maximum likelihood estimation; Mean square error methods; Principal component analysis; Self organizing feature maps;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Information Processing, 2002. ICONIP '02. Proceedings of the 9th International Conference on

Print_ISBN

981-04-7524-1

Type

conf

DOI

10.1109/ICONIP.2002.1202840

Filename

1202840