A feature extraction algorithm based on the Hilbert-Schmidt independence criterion

Authors:

Povilas DANIUŠIS, Pranas VAITKUS

AMS Classification:

62H20, 62H30

Keywords:

Hilbert-Schmidt independence criterion (HSIC), feature extraction, neural networks

Abstract:

In this article, we propose a non-linear feature extraction algorithm based on the Hilbert-Schmidt independence criterion (HSIC). The proposed algorithm was tested on several real-world benchmark data sets.

Download paper:

Daniusis&Vaitkus-09.pdf

Vol. 4 (12), 2009