We study parametric unsupervised mixture learning. We measure the loss of intrinsic information from the observations to complex mixture models, and then to simple mixture models. We present a geometric picture, where all these representations are regarded as free points in the space of probability distributions. Based on minimum description length, we derive a simple geometric principle to learn all these models together. We present a new learning machine with theories, algorithms, and simulations.
Details
Title
Information geometry and minimum description length networks
Author(s)
Sun, Ke ( Computer Vision and Multimedia Laboratory, University of Geneva, Switzerland) Wang, Jun ( Expedia, Switzerland) Kalousis, Alexandros ( Haute école de gestion de Genève, HES-SO Haute Ecole Spécialisée de Suisse Occidentale) Marchand-Maillet, Stéphane ( Computer Vision and Multimedia Laboratory, University of Geneva, Switzerland )
Date
2015-07
Published in
Journal of machine learning research : proceedings of the 32nd International Conference on Machine Learning, pp. 49–58, 2015
Volume
2015, vol. 37, pp. 49–58
Publisher
[S.l.] , Journal of machine learning research
Pagination
10 p.
Presented at
32nd International Conference on Machine Learning, Lille, France, 06/07/2015 / 11/07/2015
ISSN
1938-7228
Paper type
full paper
Faculty
Economie et Services
School
HEG - Genève
Institute
CRAG - Centre de Recherche Appliquée en Gestion