Kernel absolute summability is only sufficient for RKHS stability.
Mauro BisiaccoGianluigi PillonettoPublished in: CoRR (2019)
Keyphrases
- reproducing kernel hilbert space
- kernel methods
- kernel function
- loss function
- reproducing kernel
- special case
- gaussian kernel
- learning theory
- kernel matrix
- kernel machines
- regularized least squares
- gaussian kernels
- domain adaptation
- euclidean space
- distance measure
- density estimation
- gaussian process
- feature space
- input space
- support vector
- learning problems
- random projections
- positive definite
- representer theorem
- real valued
- hilbert space
- feature selection
- machine learning
- face recognition
- graph kernels
- kernel pca
- dimension reduction
- data dependent
- linear model