CaltechAUTHORS
  A Caltech Library Service

Discriminative Clustering by Regularized Information Maximization

Gomes, Ryan and Krause, Andreas and Perona, Pietro (2010) Discriminative Clustering by Regularized Information Maximization. In: Advances in Neural Information Processing Systems 23. Advances in Neural Information Processing Systems. No.22. Nueral Information Processing Systems , La Jolla, CA. ISBN 9781617823800. http://resolver.caltech.edu/CaltechAUTHORS:20160331-165410183

[img] PDF - Published Version
See Usage Policy.

959Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20160331-165410183

Abstract

Is there a principled way to learn a probabilistic discriminative classifier from an unlabeled data set? We present a framework that simultaneously clusters the data and trains a discriminative classifier. We call it Regularized Information Maximization (RIM). RIM optimizes an intuitive information-theoretic objective function which balances class separation, class balance and classifier complexity. The approach can flexibly incorporate different likelihood functions, express prior assumptions about the relative size of different classes and incorporate partial labels for semi-supervised learning. In particular, we instantiate the framework to unsupervised, multi-class kernelized logistic regression. Our empirical evaluation indicates that RIM outperforms existing methods on several real data sets, and demonstrates that RIM is an effective model selection method.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://papers.nips.cc/paper/4154-discriminative-clustering-by-regularized-information-maximizationOrganizationPaper
ORCID:
AuthorORCID
Perona, Pietro0000-0002-7583-5809
Additional Information:©2010 Neural Information Processing Systems. We thank Alex Smola for helpful comments and discussion, and Thanos Siapas for providing the neural tetrode data. This research was partially supported by NSF grant IIS-0953413, a gift from Microsoft Corporation, and ONR MURI Grant N00014-06-1-0734.
Funders:
Funding AgencyGrant Number
NSFIIS-0953413
Microsoft CorporationUNSPECIFIED
Office of Naval Research (ONR)N00014-06-1-0734
Record Number:CaltechAUTHORS:20160331-165410183
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20160331-165410183
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:65825
Collection:CaltechAUTHORS
Deposited By: Kristin Buxton
Deposited On:01 Apr 2016 19:32
Last Modified:08 May 2017 21:52

Repository Staff Only: item control page