Gaussian Process Topic Models

Amrudin Agovic, Arindam Banerjee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Scopus citations

Abstract

We introduce Gaussian Process Topic Models (GPTMs), a new family of topic models which can leverage a kernel among documents while extracting correlated topics. GPTMs can be considered a systematic generalization of the Correlated Topic Models (CTMs) using ideas from Gaussian Process (GP) based embedding. Since GPTMs work with both a topic covariance matrix and a document kernel matrix, learning GPTMs involves a novel component-solving a suitable Sylvester equation capturing both topic and document dependencies. The efficacy of GPTMs is demonstrated with experiments evaluating the quality of both topic modeling and embedding.

Original languageEnglish (US)
Title of host publicationProceedings of the 26th Conference on Uncertainty in Artificial Intelligence, UAI 2010
Pages10-19
Number of pages10
StatePublished - Dec 1 2010
Event26th Conference on Uncertainty in Artificial Intelligence, UAI 2010 - Catalina Island, CA, United States
Duration: Jul 8 2010Jul 11 2010

Publication series

NameProceedings of the 26th Conference on Uncertainty in Artificial Intelligence, UAI 2010

Other

Other26th Conference on Uncertainty in Artificial Intelligence, UAI 2010
CountryUnited States
CityCatalina Island, CA
Period7/8/107/11/10

Fingerprint Dive into the research topics of 'Gaussian Process Topic Models'. Together they form a unique fingerprint.

Cite this