An efficient method for computing leave-one-out error in support vector machines with Gaussian kernels.
Martin M. S. LeeS. Sathiya KeerthiChong Jin OngDennis DeCostePublished in: IEEE Trans. Neural Networks (2004)
Keyphrases
- gaussian kernels
- efficient method for computing
- multiple kernel learning
- learning rate
- kernel function
- gaussian kernel
- feature space
- general loss functions
- high dimensional feature space
- reproducing kernel hilbert space
- kernel learning
- linear combination
- kernel methods
- convergence rate
- machine learning
- least squares
- image analysis
- binary classification
- learning problems
- semi supervised learning
- learning algorithm