Robust ICA for super-Gaussian sources

Frank C. Meinecke, Stefan Harmeling, Klaus Muller

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Most ICA algorithms are sensitive to outliers. Instead of robustifying existing algorithms by outlier rejection techniques, we show how a simple outlier index can be used directly to solve the ICA problem for super-Gaussian source signals. This ICA method is outlier-robust by construction and can be used for standard ICA as well as for overcomplete ICA (i.e. more source signals than observed signals (mixtures)).

Original languageEnglish
Pages (from-to)217-224
Number of pages8
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3195
Publication statusPublished - 2004 Dec 1
Externally publishedYes

Fingerprint

Independent component analysis
Outlier
Rejection

ASJC Scopus subject areas

  • Computer Science(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Theoretical Computer Science

Cite this

Robust ICA for super-Gaussian sources. / Meinecke, Frank C.; Harmeling, Stefan; Muller, Klaus.

In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 3195, 01.12.2004, p. 217-224.

Research output: Contribution to journalArticle

@article{59a7624cce5640b8b80f4ffe524a42e5,
title = "Robust ICA for super-Gaussian sources",
abstract = "Most ICA algorithms are sensitive to outliers. Instead of robustifying existing algorithms by outlier rejection techniques, we show how a simple outlier index can be used directly to solve the ICA problem for super-Gaussian source signals. This ICA method is outlier-robust by construction and can be used for standard ICA as well as for overcomplete ICA (i.e. more source signals than observed signals (mixtures)).",
author = "Meinecke, {Frank C.} and Stefan Harmeling and Klaus Muller",
year = "2004",
month = "12",
day = "1",
language = "English",
volume = "3195",
pages = "217--224",
journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Robust ICA for super-Gaussian sources

AU - Meinecke, Frank C.

AU - Harmeling, Stefan

AU - Muller, Klaus

PY - 2004/12/1

Y1 - 2004/12/1

N2 - Most ICA algorithms are sensitive to outliers. Instead of robustifying existing algorithms by outlier rejection techniques, we show how a simple outlier index can be used directly to solve the ICA problem for super-Gaussian source signals. This ICA method is outlier-robust by construction and can be used for standard ICA as well as for overcomplete ICA (i.e. more source signals than observed signals (mixtures)).

AB - Most ICA algorithms are sensitive to outliers. Instead of robustifying existing algorithms by outlier rejection techniques, we show how a simple outlier index can be used directly to solve the ICA problem for super-Gaussian source signals. This ICA method is outlier-robust by construction and can be used for standard ICA as well as for overcomplete ICA (i.e. more source signals than observed signals (mixtures)).

UR - http://www.scopus.com/inward/record.url?scp=35048843771&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35048843771&partnerID=8YFLogxK

M3 - Article

VL - 3195

SP - 217

EP - 224

JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SN - 0302-9743

ER -