Model Selection as a Multiple Testing Procedure: Improving Akaike's Information Criterion.
Adrien SaumardFabien NavarroPublished in: CoRR (2018)
Keyphrases
- information criterion
- model selection
- cross validation
- marginal likelihood
- automatic model selection
- bayesian information criterion
- model selection criteria
- hyperparameters
- probability model
- regression model
- sample size
- parameter estimation
- leave one out cross validation
- selection criterion
- generalization error
- machine learning
- mixture model
- factor analysis
- similarity measure
- support vector machine
- variable selection
- bayesian networks
- feature extraction
- feature selection
- worst case