High-Dimensional Kernel Methods under Covariate Shift: Data-Dependent Implicit Regularization.
Yihang ChenFanghui LiuTaiji SuzukiVolkan CevherPublished in: CoRR (2024)
Keyphrases
- data dependent
- kernel methods
- high dimensional
- rademacher complexity
- kernel function
- feature space
- reproducing kernel hilbert space
- learning problems
- low dimensional
- input space
- kernel matrix
- training and test data
- support vector
- similarity search
- risk bounds
- machine learning
- generalization bounds
- dimensionality reduction
- high dimensional data
- data points
- support vector machine
- feature selection
- training samples
- domain adaptation
- learning tasks
- kernel learning
- energy functional
- nearest neighbor
- feature vectors
- support vectors
- data distribution
- image processing