Discriminative Model Selection for Density Models

Bo Thiesson, Christopher Meek
Proceedings of the Ninth International Workshop on Artificial Intelligence and Statistics, PMLR R4:270-275, 2003.

Abstract

Density models are a popular tool for building classifiers. When using density models to build a classifier, one typically learns a separate density model for each class of interest. These density models are then combined to make a classifier through the use of Bayes’ rule utilizing the prior distribution over the classes. In this paper, we provide a discriminative method for choosing among alternative density models for each class to improve classification accuracy.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR4-thiesson03a, title = {Discriminative Model Selection for Density Models}, author = {Thiesson, Bo and Meek, Christopher}, booktitle = {Proceedings of the Ninth International Workshop on Artificial Intelligence and Statistics}, pages = {270--275}, year = {2003}, editor = {Bishop, Christopher M. and Frey, Brendan J.}, volume = {R4}, series = {Proceedings of Machine Learning Research}, month = {03--06 Jan}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/r4/thiesson03a/thiesson03a.pdf}, url = {https://proceedings.mlr.press/r4/thiesson03a.html}, abstract = {Density models are a popular tool for building classifiers. When using density models to build a classifier, one typically learns a separate density model for each class of interest. These density models are then combined to make a classifier through the use of Bayes’ rule utilizing the prior distribution over the classes. In this paper, we provide a discriminative method for choosing among alternative density models for each class to improve classification accuracy.}, note = {Reissued by PMLR on 01 April 2021.} }
Endnote
%0 Conference Paper %T Discriminative Model Selection for Density Models %A Bo Thiesson %A Christopher Meek %B Proceedings of the Ninth International Workshop on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2003 %E Christopher M. Bishop %E Brendan J. Frey %F pmlr-vR4-thiesson03a %I PMLR %P 270--275 %U https://proceedings.mlr.press/r4/thiesson03a.html %V R4 %X Density models are a popular tool for building classifiers. When using density models to build a classifier, one typically learns a separate density model for each class of interest. These density models are then combined to make a classifier through the use of Bayes’ rule utilizing the prior distribution over the classes. In this paper, we provide a discriminative method for choosing among alternative density models for each class to improve classification accuracy. %Z Reissued by PMLR on 01 April 2021.
APA
Thiesson, B. & Meek, C.. (2003). Discriminative Model Selection for Density Models. Proceedings of the Ninth International Workshop on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research R4:270-275 Available from https://proceedings.mlr.press/r4/thiesson03a.html. Reissued by PMLR on 01 April 2021.

Related Material