RFN: A Random-Feature Based Newton Method for Empirical Risk Minimization in Reproducing Kernel Hilbert Spaces.
Ting-Jui ChangShahin ShahrampourPublished in: IEEE Trans. Signal Process. (2022)
Keyphrases
- uniform convergence
- empirical risk minimization
- reproducing kernel hilbert space
- sufficient conditions
- learning rate
- kernel methods
- risk minimization
- real valued
- loss function
- generalization error
- learning theory
- vc dimension
- data dependent
- distance measure
- euclidean space
- domain adaptation
- input space
- learning problems
- kernel function
- special case
- density estimation
- gaussian process
- pairwise
- metric learning
- learning algorithm