Managing the computational cost of model selection and cross-validation in extreme learning machines via Cholesky, SVD, QR and eigen decompositions.
Yiannis KokkinosKonstantinos G. MargaritisPublished in: Neurocomputing (2018)
Keyphrases
- model selection
- cross validation
- singular value decomposition
- extreme learning machines
- principal component analysis
- kernel matrix
- hyperparameters
- variable selection
- generalization error
- low rank
- least squares
- dimensionality reduction
- parameter estimation
- sample size
- information criterion
- dimension reduction
- regression model
- selection criterion
- meta learning
- statistical inference
- error estimation
- mixture model
- model selection criteria
- leave one out cross validation
- gaussian process
- machine learning
- feature selection
- face recognition
- density estimation
- latent variables
- closed form
- feature space