Powered Hawkes-Dirichlet Process: Challenging Textual Clustering using a Flexible Temporal Prior.
Gaël Poux-MédardJulien VelcinSabine LoudcherPublished in: CoRR (2021)
Keyphrases
- dirichlet process
- dirichlet process mixture models
- bayesian model
- mixture model
- multi task learning
- gaussian processes
- chinese restaurant process
- clustering algorithm
- sampling algorithm
- k means
- latent topics
- unsupervised learning
- posterior distribution
- prior knowledge
- markov chain monte carlo
- bayesian framework
- bayesian inference
- generative model
- prior information
- variational methods
- learning tasks
- variational inference
- probabilistic model
- density estimation
- monte carlo
- maximum likelihood
- language model
- multi class
- maximum a posteriori