An algebraic characterization of the optimum of regularized kernel methods.
Francesco DinuzzoGiuseppe De NicolaoPublished in: Mach. Learn. (2009)
Keyphrases
- kernel methods
- kernel function
- learning problems
- support vector
- feature space
- machine learning
- support vector machine
- kernel matrix
- reproducing kernel hilbert space
- learning tasks
- kernel trick
- least squares
- similarity function
- high dimensional feature space
- kernel pca
- singular value decomposition
- multiple kernel
- objective function
- kernel matrices
- kernel ridge regression
- reproducing kernel
- neural network
- positive semidefinite
- feature extraction
- decision trees
- data mining