Parameter estimation in conditionally Gaussian pairwise Markov switching models and unsupervised smoothing.
Fei ZhengStéphane DerrodeWojciech PieczynskiPublished in: MLSP (2016)
Keyphrases
- parameter estimation
- maximum likelihood
- markov random field
- statistical models
- pairwise
- markov fields
- random fields
- em algorithm
- model selection
- least squares
- model fitting
- parameter values
- parameter estimation algorithm
- parameter estimates
- approximate inference
- semi supervised
- estimation problems
- belief propagation
- posterior distribution
- expectation maximization
- experimental data
- maximum likelihood estimation
- parameters estimation
- similarity measure
- unsupervised image segmentation
- image processing
- machine learning