Title :
Non-parametric dependent components
Author :
Klami, Arto ; Kaski, Samuel
Author_Institution :
Dept. of Comput. Sci., Helsinki Univ., Finland
Abstract :
Canonical correlation analysis (CCA) is equivalent to finding mutual information-maximizing projections for normally distributed data. We remove the restriction of normality by non-parametric estimation, and formulate the problem of finding dependent components with a connection to Bayes factors. The method is applied for characterizing yeast stress by finding what is in common in several different stress conditions.
Keywords :
Bayes methods; correlation methods; normal distribution; stress analysis; Bayes factors; CCA; canonical correlation analysis; mutual information-maximizing projections; nonparametric dependent components; nonparametric estimation; normality restriction; normally distributed data; stress conditions; yeast stress; Bayesian methods; Computer science; Cost function; Distributed computing; Eigenvalues and eigenfunctions; Fungi; Information analysis; Mutual information; Stress;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on
Print_ISBN :
0-7803-8874-7
DOI :
10.1109/ICASSP.2005.1416277
