1UBC-Computer Science - Computer Science Department (UBC DEPARTMENT OF COMPUTER SCIENCE ICICS/CS Building 201-2366 Main Mall Vancouver, B.C. V6T 1Z4 General Enquiries: Tel: 604-822-3061 E-mail: info@cs.ubc.ca Fax: 604-822-5485 - Canada)
UBC - University of British Columbia (Vancouver Campus, , 2329 West Mall, Vancouver, BC Canada V6T 1Z4 -
Okanagan Campus, 3333 University Way, Kelowna, BC Canada V1V 1V7 - Canada)
2Statistics - Department of Statistics (DEPARTMENT OF STATISTICS 333-6356 Agricultural Road University of British Columbia Vancouver, BC, V6T 1Z2 Tel: 604 822 0570 Fax: 604 822 6960 - Canada)
UBC - University of British Columbia (Vancouver Campus, , 2329 West Mall, Vancouver, BC Canada V6T 1Z4 -
Okanagan Campus, 3333 University Way, Kelowna, BC Canada V1V 1V7 - Canada)
Abstract : We present a method for variable selection/weighting in an unsupervised learning context using Bayesian shrinkage. The basis for the model parameters and cluster assignments can be computed simultaneous using an efficient EM algorithm. Applying our Bayesian shrinkage model to a complex problem in object recognition (Duygulu, Barnard, de Freitas and Forsyth 2002), our experiments yied good results.
https://hal.inria.fr/inria-00548235
Contributeur : Thoth Team
<>
Soumis le : lundi 20 décembre 2010 - 14:51:34
Dernière modification le : lundi 20 décembre 2010 - 15:28:45
Document(s) archivé(s) le : lundi 21 mars 2011 - 02:46:41
Peter Carbonetto, Nando De Freitas, Paul Gustafson, Natalie Thompson. Bayesian feature weighting for unsupervised learning, with application to object recognition. Christopher M. Bishop and Brendan J. Frey. Artificial Intelligence and Statistics (AI & Statistics '03), Jan 2003, Key West, United States. Society for Artificial Intelligence and Statistics, 2003, 〈http://research.microsoft.com/en-us/um/cambridge/events/aistats2003/proceedings/papers.htm〉. 〈inria-00548235〉
