stat.MLJun 19, 2026

Kernel Ridge Regression over Hölder Classes: Function and Derivative Losses

Authors: Yuxuan Hou, Yicheng Li

Abstract

We study ordinary kernel ridge regression over full Hölder-Zygmund balls, measuring errors in Sobolev norms and in pointwise and uniform losses for mixed derivatives. For targets of smoothness ss, every RKHS norm equivalent to Hs+d/2H^{s+d/2} gives the optimal HrH^r risk for 0≤r<s0\le r<s; integer choices of rr include integrated squared derivative losses. Pointwise and uniform guarantees require additional kernel structure. A localized wavelet kernel attains the sharp derivative rates with the same regularization orders as for function estimation, whereas equivalent changes of the RKHS norm can destroy the population bias rates. For the ordinary Gaussian kernel with a shrinking bandwidth, we prove a uniform derivative risk bound with the optimal polynomial rate and an explicit, unoptimized logarithmic factor. The results allow bounded measurable design densities bounded away from zero and conditionally sub-Gaussian noise.

Explore similar work

CardsList
  1. Highly Adaptive Ridge

    Oct 3, 2024Alejandro Schuler, Mingxun Wang, Alexander Hagemeister +1Nonparametric RegressionRidge Regression

  2. Optimal Confidence Band for Kernel Gradient Flow Estimator

    May 7, 2026Yuqian Cheng, Zhuo Chen, Qian LinKernel RegressionKernel Methods