@inproceedings{219967de16424fd1a91f83cd9bfc9201,
title = "Overcomplete ICA with a geometric algorithm",
abstract = "We present an independent component analysis (ICA) algorithm based on geometric considerations [10] [11] to decompose a linear mixture of more sources than sensor signals. Bofill and Zibulevsky [2] recently proposed a two-step approach for the separation: first learn the mixing matrix, then recover the sources using a maximum-likelihood approach. We present an efficient method for the matrix-recovery step mimicking the standard geometric algorithm thus generalizing Bofill and Zibulevsky's method.",
author = "Theis, {Fabian J.} and Lang, {Elmar W.} and Tobias Westenhuber and Puntonet, {Carlos G.}",
year = "2002",
doi = "10.1007/3-540-46084-5_170",
language = "English",
isbn = "9783540440741",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "1049--1054",
editor = "Dorronsoro, {Jose R.} and Dorronsoro, {Jose R.}",
booktitle = "Artificial Neural Networks, ICANN 2002 - International Conference, Proceedings",
note = "2002 International Conference on Artificial Neural Networks, ICANN 2002 ; Conference date: 28-08-2002 Through 30-08-2002",
}