Context-Specific Independence in Directed Relational Probabilistic Models and its Influence on the Efficiency of Gibbs Sampling.
Daan FierensPublished in: ECAI (2010)
Keyphrases
- context specific independence
- probabilistic model
- gibbs sampling
- expectation maximization
- probabilistic inference
- topic models
- approximate inference
- bayesian networks
- graphical models
- exact inference
- belief networks
- generative model
- latent dirichlet allocation
- em algorithm
- naive bayes models
- latent variables
- parameter estimation
- markov chain monte carlo
- conditional probabilities
- bayesian inference
- markov chain
- conditional random fields
- conditional probability tables
- hidden variables
- belief propagation
- conditional independencies
- bayesian framework
- markov networks
- probabilistic graphical models
- state space
- random variables
- data sets