Predicting the difficulty of pure, strict, epistatic models: metrics for simulated model selection.
Ryan J. UrbanowiczJeff KiralisJonathan M. FisherJason H. MoorePublished in: BioData Min. (2012)
Keyphrases
- model selection
- cross validation
- parameter estimation
- hyperparameters
- model selection criteria
- sample size
- statistical inference
- selection criterion
- regression model
- bayesian learning
- statistical learning
- bayesian approaches
- mixture model
- meta learning
- error estimation
- information criterion
- generalization error
- gaussian process
- motion segmentation
- leave one out cross validation
- feature selection
- marginal likelihood
- parameter determination
- bayesian methods
- variable selection
- least squares
- machine learning
- approximate inference
- prior information
- closed form
- hypothesis tests
- support vector
- automatic model selection