Electrical Engineering
      and Computer Sciences

Electrical Engineering and Computer Sciences

COLLEGE OF ENGINEERING

UC Berkeley

Kernel Independent Component Analysis

Francis R. Bach and Michael I. Jordan

EECS Department
University of California, Berkeley
Technical Report No. UCB/CSD-01-1166
November 2001

http://www.eecs.berkeley.edu/Pubs/TechRpts/2001/CSD-01-1166.pdf

We present a class of algorithms for Independent Component Analysis (ICA) which use contrast functions based on canonical correlations in a reproducing kernel Hilbert space. On the one hand, we show that our contrast functions are related to mutual information and have desirable mathematical properties as measures of statistical dependence. On the other hand, building on recent developments in kernel methods, we show that these criteria and their derivatives can be computed efficiently. Minimizing these criteria leads to flexible and robust algorithms for ICA. We illustrate with simulations involving a wide variety of source distributions, showing that our algorithms outperform many of the presently known algorithms.


BibTeX citation:

@techreport{Bach:CSD-01-1166,
    Author = {Bach, Francis R. and Jordan, Michael I.},
    Title = {Kernel Independent Component Analysis},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2001},
    Month = {Nov},
    URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/2001/5721.html},
    Number = {UCB/CSD-01-1166},
    Abstract = {We present a class of algorithms for Independent Component Analysis (ICA) which use contrast functions based on canonical correlations in a reproducing kernel Hilbert space. On the one hand, we show that our contrast functions are related to mutual information and have desirable mathematical properties as measures of statistical dependence. On the other hand, building on recent developments in kernel methods, we show that these criteria and their derivatives can be computed efficiently. Minimizing these criteria leads to flexible and robust algorithms for ICA. We illustrate with simulations involving a wide variety of source distributions, showing that our algorithms outperform many of the presently known algorithms.}
}

EndNote citation:

%0 Report
%A Bach, Francis R.
%A Jordan, Michael I.
%T Kernel Independent Component Analysis
%I EECS Department, University of California, Berkeley
%D 2001
%@ UCB/CSD-01-1166
%U http://www.eecs.berkeley.edu/Pubs/TechRpts/2001/5721.html
%F Bach:CSD-01-1166