Header logo is ei

Analysing ICA component by injection noise

2003

Conference Paper

ei


Usually, noise is considered to be destructive. We present a new method that constructively injects noise to assess the reliability and the group structure of empirical ICA components. Simulations show that the true root-mean squared angle distances between the real sources and some source estimates can be approximated by our method. In a toy experiment, we see that we are also able to reveal the underlying group structure of extracted ICA components. Furthermore, an experiment with fetal ECG data demonstrates that our approach is useful for exploratory data analysis of real-world data.

Author(s): Harmeling, S. and Meinecke, F. and Müller, K-R.
Book Title: ICA 2003
Journal: Proceedings of the 4th International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003)
Pages: 149-154
Year: 2003
Month: April
Day: 0
Editors: Amari, S.-I. , A. Cichocki, S. Makino, N. Murata

Department(s): Empirical Inference
Bibtex Type: Conference Paper (inproceedings)

Event Name: 4th International Symposium on Independent Component Analysis and Blind Signal Separation
Event Place: Nara, Japan

Digital: 0
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inproceedings{6358,
  title = {Analysing ICA component by injection noise},
  author = {Harmeling, S. and Meinecke, F. and M{\"u}ller, K-R.},
  journal = {Proceedings of the 4th International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003)},
  booktitle = {ICA 2003},
  pages = {149-154},
  editors = {Amari, S.-I. , A. Cichocki, S. Makino, N. Murata},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  month = apr,
  year = {2003},
  month_numeric = {4}
}