Abstract
Under mild assumptions on the kernel, we obtain the best known error rates in a regularized learning scenario taking place in the corresponding reproducing kernel Hilbert space (RKHS). The main novelty in the analysis is a proof that one can use a regularization term that grows significantly slower than the standard quadratic growth in the RKHS norm.
Citation
Shahar Mendelson. Joseph Neeman. "Regularization in kernel learning." Ann. Statist. 38 (1) 526 - 565, February 2010. https://doi.org/10.1214/09-AOS728
Information