Title :
A simple ICA algorithm for non-differentiable contrasts
Author :
Lee, John A. ; Vrins, Frederic ; Verleysen, Michel
Author_Institution :
Machine Learning Group, Univ. catholique de Louvain, Louvain-la-Neuve, Belgium
Abstract :
A general-purpose algorithm is proposed for Independent Component Analysis. This algorithm is specifically designed in order to handle non-differentiable contrast functions. Sources are extracted one at a time (deflation approach). Examples are given for recently published contrast functions, e.g. the support width of the estimated source distribution.
Keywords :
independent component analysis; signal processing; ICA algorithm; estimated source distribution; general-purpose algorithm; independent component analysis; nondifferentiable contrasts; Algorithm design and analysis; Approximation algorithms; Approximation methods; Histograms; Principal component analysis; Robustness; Signal processing algorithms;
Conference_Titel :
Signal Processing Conference, 2005 13th European
Conference_Location :
Antalya
Print_ISBN :
978-160-4238-21-1
