TY - GEN
T1 - Gaussian Process Topic Models
AU - Agovic, Amrudin
AU - Banerjee, Arindam
PY - 2010/12/1
Y1 - 2010/12/1
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=80053167768&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80053167768&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:80053167768
SN - 9780974903965
T3 - Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, UAI 2010
SP - 10
EP - 19
BT - Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, UAI 2010
T2 - 26th Conference on Uncertainty in Artificial Intelligence, UAI 2010
Y2 - 8 July 2010 through 11 July 2010
ER -