Lepskii Principle in Supervised Learning - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Preprints, Working Papers, ... Year :

Lepskii Principle in Supervised Learning

Abstract

In the setting of supervised learning using reproducing kernel methods, we propose a data-dependent regularization parameter selection rule that is adaptive to the unknown regularity of the target function and is optimal both for the least-square (prediction) error and for the reproducing kernel Hilbert space (reconstruction) norm error. It is based on a modified Lepskii balancing principle using a varying family of norms.

Dates and versions

hal-02974206 , version 1 (21-10-2020)

Identifiers

Cite

Gilles Blanchard, Peter Mathé, Nicole Mücke. Lepskii Principle in Supervised Learning. 2020. ⟨hal-02974206⟩
27 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More