Supervised Hierarchical Dirichlet Processes with Variational Inference

Cheng Zhang, Carl Henrik Ek, Xavi Gratal, Florian T. Pokorny, Hedvig Kjellström
In IEEE ICCV Workshop on Inference for Probabilistic Graphical Models, 2013

Abstract

We present an extension to the Hierarchical Dirichlet Process (HDP), which allows for the inclusion of supervision. Our model marries the non-parametric benefits of HDP with those of Supervised Latent Dirichlet Allocation (SLDA) to enable learning the topic space directly from data while simultaneously including the labels within the model. The proposed model is learned using variational inference which allows for the efficient use of a large training dataset. We also present the online version of variational inference, which makes the method scalable to very large datasets. We show results comparing our model to a traditional supervised parametric topic model, SLDA, and show that it outperforms SLDA on a number of benchmark datasets.

Files

Download this publication

Bibtex

@article{zhang2013a, title = {Supervised Hierarchical Dirichlet Processes with Variational Inference}, author = {Zhang, Cheng and Ek, Carl Henrik and Gratal, Xavi and Pokorny, Florian T. and Kjellstr{\"o}m, Hedvig}, journal = {IEEE ICCV Workshop on Inference for Probabilistic Graphical Models}, year = {2013}, url = {http://www.csc.kth.se/~fpokorny/static/publications/zhang2013a.pdf}, }