Reparameterization strategies for hidden Markov models and Bayesian approaches to maximum likelihood estimation.
Christian P. RobertD. M. TitteringtonPublished in: Stat. Comput. (1998)
Keyphrases
- hidden markov models
- maximum likelihood estimation
- bayesian approaches
- minimum classification error
- maximum likelihood
- em algorithm
- expectation maximization
- gaussian processes
- parameter estimation
- probability distribution
- bayesian methods
- machine learning
- conditional random fields
- model selection
- density function
- gaussian process
- graphical models
- least squares
- discriminative training
- clustering algorithm