A criterion for model selection in the presence of incomplete data based on Kullback's symmetric divergence.
Abd-Krim SeghouaneMaïza BekaraGilles FleuryPublished in: Signal Process. (2005)
Keyphrases
- incomplete data
- model selection
- selection criterion
- model selection criteria
- hyperparameters
- marginal likelihood
- cross validation
- learning bayesian networks
- missing values
- feature selection
- missing data
- bayesian networks
- sample size
- information criterion
- bayesian information criterion
- parameter estimation
- em algorithm
- machine learning
- incomplete data sets
- mixture model
- gaussian process
- variable selection
- selection criteria
- posterior distribution
- bayesian methods
- data sets