Decoupling Sparsity and Smoothness in the Dirichlet Variational Autoencoder Topic Model.
Sophie BurkhardtStefan KramerPublished in: J. Mach. Learn. Res. (2019)
Keyphrases
- topic models
- topic modeling
- latent dirichlet allocation
- generative model
- text mining
- text documents
- image segmentation
- probabilistic model
- co occurrence
- em algorithm
- prior information
- gibbs sampling
- latent variables
- high dimensional
- sparse representation
- latent topics
- variational methods
- mixture model
- variational inference
- optical flow
- baseline models
- microblog posts
- parameter estimation
- probabilistic topic models
- knowledge representation
- latent topic model
- learning algorithm
- probabilistic latent semantic analysis
- feature selection