Model Selection and Overfitting in Genetic Programming: Empirical Study.
Jan ZegklitzPetr PosíkPublished in: CoRR (2015)
Keyphrases
- empirical studies
- model selection
- genetic programming
- cross validation
- generalization error
- hyperparameters
- parameter estimation
- fitness function
- variable selection
- sample size
- regression model
- bayesian learning
- evolutionary algorithm
- empirical analysis
- information criterion
- statistical learning
- statistical inference
- machine learning
- mixture model
- genetic algorithm
- error estimation
- motion segmentation
- meta learning
- decision trees
- feature selection
- selection criterion
- leave one out cross validation
- data sets
- learning machines
- marginal likelihood
- unsupervised learning
- generative model
- support vector
- gaussian process