Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

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

Summary: Soft dimension reduction for ICA by joint
diagonalization on the Stiefel manifold
Fabian J. Theis1
, Thomas P. Cason2
, and P.-A. Absil2
1
CMB, Institute of Bioinformatics and Systems Biology,
Helmholtz Zentrum M¨unchen, Germany, and
MPI for Dynamics and Self-Organization, G¨ottingen, Germany
fabian.theis@helmholtz-muenchen.de, http://cmb.helmholtz-muenchen.de
2
Department of Mathematical Engineering, Universit´e catholique de Louvain,
B-1348 Louvain-la-Neuve, Belgium (http://www.inma.ucl.ac.be/~{cason,absil})
Abstract. Joint diagonalization for ICA is often performed on the or-
thogonal 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 Rn
. 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-

  

Source: Absil, Pierre-Antoine - Département d'ingénierie Mathématique, Université Catholique de Louvain

 

Collections: Mathematics