| HAL-Inria | Publications, software ... of Inria's scientists |
Conference papers
Non-linear feature extraction by the coordination of mixture models
Abstract : We present a method for non-linear data projection that offers non-linear versions of Principal Component Analysis and Canonical Correlation Analysis. The data is accessed through a probabilistic mixture model only, therefore any mixture model for any type of data can be plugged in. Gaussian mixtures are one example, but mixtures of Bernoulli's to model discrete data might be used as well. The algorithm minimizes an objective function that exhibits one global optimum that can be found by finding the eigenvectors of some matrix. Experimental results on toy data and real data are provided.
Cited literature [15 references]
https://hal.inria.fr/inria-00321490
Contributor : Jakob Verbeek <>
Submitted on : Tuesday, March 8, 2011 - 3:04:10 PM
Last modification on : Monday, September 25, 2017 - 10:08:04 AM
Document(s) archivé(s) le : Thursday, June 9, 2011 - 2:45:01 AM
Contributor : Jakob Verbeek <>
Submitted on : Tuesday, March 8, 2011 - 3:04:10 PM
Last modification on : Monday, September 25, 2017 - 10:08:04 AM
Document(s) archivé(s) le : Thursday, June 9, 2011 - 2:45:01 AM
Files
verbeek03asci.pdf
Files produced by the author(s)