PuSH - Publication Server of Helmholtz Zentrum München

Theis, F.J. ; Cason, T.P.* ; Absil, P.A.*

Soft dimension reduction for ICA by joint diagonalization on the Stiefel manifold.

In: Independent Component Analysis and Signal Separation. Berlin [u.a.]: Springer, 2009. 354-361 (Lect. Notes Comput. Sc. ; 5441)
DOI Order publishers version
Open Access Green as soon as Postprint is submitted to ZB.
Joint diagonalization for ICA is often performed on the orthogonal group after a pre-whitening step. Here we assume that we only want to extract a few sources after pre-whitening, and hence work on the Stiefel manifold of $p$-frames in $R^n$. The resulting method does not only use second-order statistics to estimate the dimension reduction and is therefore denoted as soft dimension reduction. We employ a trust-region method for minimizing the cost function on the Stiefel manifold. Applications to a toy example and functional MRI data show a higher numerical efficiency, especially when $p$ is much smaller than $n$, and more robust performance in the presence of strong noise than methods based on pre-whitening.
Altmetric
Additional Metrics?
Tags
Icb_ML
Edit extra informations Login
Publication type Article: Edited volume or book chapter
Editors Adali, T.*
Keywords independent component analysis (ICA); blind source separation (BSS); joint diagonalization; soft dimension reduction
ISSN (print) / ISBN 0302-9743
e-ISSN 1611-3349
Book Volume Title Independent Component Analysis and Signal Separation
Quellenangaben Volume: 5441, Issue: , Pages: 354-361 Article Number: , Supplement: ,
Publisher Springer
Publishing Place Berlin [u.a.]