General Description
This software implementation included here is related to our paper presented at a conference [GaKi05]. It evaluates the sensitivity to outliers of the following five independent component analysis (ICA) algorithms:
- FastICA;
- Extended-Infomax;
- JADE;
- Radical-ICA; and
- Beta-divergence.
The sensitivity to outliers is evaluated using the following measures:
- The Amari separation performance index;
- The optimum angle of rotation error; and
- The contrast function difference.
The signal to be demixed by the ICA is a simulated outlier-contaminated mixture of sources. This zip file contains the Matlab code used to generate the results for that paper.
Reference
- [GaKi05]
- N. Gadhok and W. Kinsner, "Rotation sensitivity of independent component analysis to outliers," in Proc. IEEE 2005 Canadian Conference on Electrical and Computer Engineering, CCECE05 (Saskatoon, Saskatchwan; May 1-4, 2005) pp. 1419-1424 (6 pp.), 2005.
Code Description
File Name
- CCECE05_ICA_Code_v10.zip
(Contains 11 .m files and 6 subdirectories, and 1 readme file)
Version
- 1.0
Subdirectories
- FastICA (19 .m files and 1 zip file)
- Extended-Infomax (6 .m files)
- JADE (5 .m files)
- Radical-ICA (7 .m files, 1 readme, example subdirectory, and 1 .tar file)
- Beta-divergence (10 .m files)
Notes
- Internal Versions: Each .m file has a version listed with the file.
- CCECE05_Exp1.m is the primary caller program for generating the results contained in [GaKi05].
- View the readme.txt file for details on how to run the program.
- A MATLAB implementation of beta-divergence for BSS is provided.
Feedback
- Please provide any feedback.
Legal Notes
- This software is provided for research purposes only.
- No commercial use in any form is permitted.
- WE MAKE NO WARRANTIES, EITHER EXPRESS OR IMPLIED, REGARDING THE ENCLOSED COMPUTER SOFTWARE PACKAGE.
Code Downloading
Unzip the file after downloading.
- CCECE05_ICA_Code_v10.zip (zip; 306 KB)
|