On the Benefits of Large Learning Rates for Kernel Methods.
Gaspard BeugnotJulien MairalAlessandro RudiPublished in: CoRR (2022)
Keyphrases
- kernel methods
- learning rate
- gaussian kernels
- kernel function
- reproducing kernel hilbert space
- gaussian kernel
- learning problems
- convergence rate
- machine learning
- high dimensional feature space
- learning algorithm
- support vector
- feature space
- kernel matrix
- support vector machine
- multiple kernel learning
- learning tasks
- convergence speed
- kernel learning
- upper bound
- discriminant analysis
- kernel pca