On the Problem in Model Selection of Neural Network Regression in Overrealizable Scenario.
Katsuyuki HagiwaraKazuhiro KunoShiro UsuiPublished in: IJCNN (6) (2000)
Keyphrases
- model selection
- neural network
- regression model
- cross validation
- hyperparameters
- parameter estimation
- bayesian learning
- sample size
- statistical learning
- selection criterion
- statistical inference
- machine learning
- error estimation
- meta learning
- variable selection
- generalization error
- artificial neural networks
- feature selection
- marginal likelihood
- gaussian process
- back propagation
- mixture model
- automatic model selection
- leave one out cross validation
- model selection criteria
- bayesian methods
- real world
- generalization bounds
- multi layer perceptron
- bp neural network
- semi supervised