Variational Bayes for estimating the parameters of a hidden Potts model.
Clare A. McGroryD. M. TitteringtonR. ReevesAnthony N. PettittPublished in: Stat. Comput. (2009)
Keyphrases
- variational bayes
- potts model
- hyperparameters
- prior distribution
- posterior distribution
- bayesian inference
- maximum a posteriori
- cross validation
- markov random field
- latent variables
- max flow
- parameter estimation
- gaussian mixture model
- model selection
- bayesian framework
- expectation maximization
- maximum likelihood
- free energy
- random sampling
- probability distribution
- closed form
- support vector
- parameter space
- cellular automata
- incremental learning
- graph cuts
- sample size
- training set