Showing 1–1 of 1 results for author: Gerschinovitz, S

Online nonparametric regression with Sobolev kernels

Authors: Oleksandr Zadorozhnyi, Pierre Gaillard, Sebastien Gerschinovitz, Alessandro Rudi

Abstract: In this work we investigate the variation of the online kernelized ridge regression algorithm in the setting of $d-$dimensional adversarial nonparametric regression. We derive the regret upper bounds on the classes of Sobolev spaces $W_{p}^β(\mathcal{X})$, $p\geq 2, β>\frac{d}{p}$. The upper bounds are supported by the minimax regret analysis, which reveals that in the cases $β> \frac{d}{2}$ or… ▽ More In this work we investigate the variation of the online kernelized ridge regression algorithm in the setting of $d-$dimensional adversarial nonparametric regression. We derive the regret upper bounds on the classes of Sobolev spaces $W_{p}^β(\mathcal{X})$, $p\geq 2, β>\frac{d}{p}$. The upper bounds are supported by the minimax regret analysis, which reveals that in the cases $β> \frac{d}{2}$ or $p=\infty$ these rates are (essentially) optimal. Finally, we compare the performance of the kernelized ridge regression forecaster to the known non-parametric forecasters in terms of the regret rates and their computational complexity as well as to the excess risk rates in the setting of statistical (i.i.d.) nonparametric regression. △ Less

Submitted 13 July, 2021; v1 submitted 6 February, 2021; originally announced February 2021.

Comments: 40 pages, 5 figures, 3 tables (version 2)