TL;DR本论文证明了当H是一个Sobolev RKHS时,对于任何0 < s < 1,KRR都是极小化最优的,解决了长期存在的问题。
Abstract
In the misspecified kernel ridge regression problem, researchers usually assume the underground true function $f_{\rho}^{*} \in [\mathcal{H}]^{s}$, a less-smooth interpolation space of a reproducing kernel hilbert space