Semi-nonnegative Independent Component Analysis: The (3,4)-SENICAexp Method.

Abstract : To solve the Independent Component Analysis (ICA) problem under the constraint of nonnegative mixture, we propose an iterative algorithm, called (3,4)-SENICAexp. This method profits from some interesting properties enjoyed by third and fourth order statistics in the presence of mixed independent processes, imposing the nonnegativity of the mixture by means of an exponential change of variable. This process allows us to obtain an unconstrained problem, optimized using an ELSALS-like procedure. Our approach is tested on synthetic magnetic resonance spectroscopic imaging data and compared to two existing ICA methods, namely SOBI and CoM2.
Type de document :
Communication dans un congrès
Proceedings of the 9th International Conference Latent Variable Analysis and Signal Separation LVA/ICA 2010, Sep 2010, St. Malo, France. Springer Berlin Heidelberg, pp.612-619, 2010, Lecture Notes in Computer Science. <10.1007/978-3-642-15995-4_76>
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00910705
Contributeur : Morgane Le Corre <>
Soumis le : jeudi 28 novembre 2013 - 10:28:02
Dernière modification le : jeudi 20 octobre 2016 - 11:53:28

Identifiants

Collections

LTSI | I3S | UNICE

Citation

Julie Coloigner, Laurent Albera, Ahmad Karfoul, Amar Kachenoura, Pierre Comon, et al.. Semi-nonnegative Independent Component Analysis: The (3,4)-SENICAexp Method.. Proceedings of the 9th International Conference Latent Variable Analysis and Signal Separation LVA/ICA 2010, Sep 2010, St. Malo, France. Springer Berlin Heidelberg, pp.612-619, 2010, Lecture Notes in Computer Science. <10.1007/978-3-642-15995-4_76>. <hal-00910705>

Partager

Métriques

Consultations de la notice

111