Flexible Kernel Independent Component Analysis Algorithm and its Local Stability on Feature Space

Author

Li, Lei

Author_Institution

Fac. of Math. & Phys., Nanjing Univ. of Posts & Telecommun.

fYear

2006

fDate

13-16 Aug. 2006

Firstpage

2990

Lastpage

2994

Abstract

In this paper a novel flexible kernel independent component analysis (FKICA) algorithm is defined and its local stability on feature space is discussed. In the FKICA algorithm, the shape of nonlinear activation function in the learning algorithm varies depending on the Gaussian exponent, which is properly selected according to the kurtosis of estimated source in feature space. In the framework of the natural gradient in Stiefel manifold, the FKICA algorithm is visited and some results about its local stability analysis are presented

Keywords

Gaussian distribution; independent component analysis; learning (artificial intelligence); nonlinear functions; Gaussian exponent; Stiefel manifold; flexible kernel independent component analysis algorithm; learning algorithm; nonlinear activation function; stability analysis; Cybernetics; Eigenvalues and eigenfunctions; Independent component analysis; Kernel; Machine learning; Machine learning algorithms; Manifolds; Mathematics; Matrix converters; Physics; Probability distribution; Stability analysis; Support vector machines; FKICA; activation function; feature space; local stability analysis; natural gradient;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2006 International Conference on

Conference_Location

Dalian, China

Print_ISBN

1-4244-0061-9

Type

conf

DOI

10.1109/ICMLC.2006.259152

Filename

4028575