A robustification of independent component analysis

By Guy Brys, Mia Hubert, Peter J. Rousseeuw. 2005

84 citations recorded by the source corpus.

Subfields: Analytical Chemistry, Statistics and Probability, Signal Processing

View Paper: https://www.semanticscholar.org/paper/2754f6d991070ac93d8a5223ca6887b82626aea0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Analytical Chemistry

1. Peter J. Rousseeuw
2. Mia Hubert
3. Ellen Vandervieren
4. S. Van Aelst
5. Klaus Nordhausen
6. David E. Tyler
7. Stephan Van der Veeken
8. K. V. Branden

In Statistics and Probability

1. Mia Hubert
2. Peter J. Rousseeuw
3. S. Van Aelst
4. Ellen Vandervieren
5. Stephan Van der Veeken
6. Klaus Nordhausen
7. David E. Tyler
8. K. V. Branden

In Signal Processing

1. Ellen Vandervieren
2. Klaus Nordhausen
3. David E. Tyler
4. Mia Hubert
5. S. Van Aelst
6. Peter J. Rousseeuw

Citations in the corpus

• High-Breakdown Robust Multivariate Methods (2008, methodology): “Robust methods were proposed for factor analysis (Pison et al., 2003) and independent component analysis ( Brys, Hubert and Rousseeuw, 2005 ).”
• A cautionary note on robust covariance plug-in methods (2014, methodology): “Problems in practice, when simply replacing the regular covariance matrix with the MCDin the context of the popular fastICA method, have been noted by Brys et al. (2005).”
• An adjusted boxplot for skewed distributions (2008, methodology): “While in this paper we focussed on the detection of univariate outliers, the idea of the adjusted boxplot has been applied to detect multivariate outliers in the context of independent component analysis (Brys et al., 2005).”
• Outlier detection for skewed data (2008, background): “The contamination was chosen to be clustered as from the simulation study in [23] this setting appeared to be the most difficult to handle.”