Name
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. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/ICCVW.2013.41 | Computer Vision Workshops |
Keywords | Field | DocType |
variational inference,slda,statistical distributions,large datasets,supervised hierarchical dirichlet process,shdp,benchmark datasets,topic model,large training dataset,inference mechanisms,learning (artificial intelligence),supervised latent dirichlet allocation,traditional supervised parametric topic,efficient use,variatioanl inference,supervised parametric topic model,hdp,topic space,supervised hierarchical dirichlet processes,hierarchical dirichlet process,learning artificial intelligence | Data mining,Data modeling,Hierarchical Dirichlet process,Latent Dirichlet allocation,Computer science,Artificial intelligence,Dirichlet distribution,Pattern recognition,Inference,Parametric statistics,Topic model,Machine learning,Scalability | Conference |
Volume | Issue | Citations |
2013 | 1 | 8 |
PageRank | References | Authors |
0.44 | 10 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| cheng zhang | 1 | 51 | 13.71 |
| carl henrik ek | 2 | 327 | 30.76 |
| Xavi Gratal | 3 | 72 | 3.04 |
| Florian T. Pokorny | 4 | 158 | 20.07 |
| hedvig kjellstrom | 5 | 491 | 42.24 |