Saved in:
Bibliographic Details
Main Authors: Della Vecchia, Andrea, Watusadisi, Arnaud Mavakala, De Vito, Ernesto, Rosasco, Lorenzo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.14083
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908371476545536
author Della Vecchia, Andrea
Watusadisi, Arnaud Mavakala
De Vito, Ernesto
Rosasco, Lorenzo
author_facet Della Vecchia, Andrea
Watusadisi, Arnaud Mavakala
De Vito, Ernesto
Rosasco, Lorenzo
contents This paper addresses the covariate shift problem in the context of nonparametric regression within reproducing kernel Hilbert spaces (RKHSs). Covariate shift arises in supervised learning when the input distributions of the training and test data differ, presenting additional challenges for learning. Although kernel methods have optimal statistical properties, their high computational demands in terms of time and, particularly, memory, limit their scalability to large datasets. To address this limitation, the main focus of this paper is to explore the trade-off between computational efficiency and statistical accuracy under covariate shift. We investigate the use of random projections where the hypothesis space consists of a random subspace within a given RKHS. Our results show that, even in the presence of covariate shift, significant computational savings can be achieved without compromising learning performance.
format Preprint
id arxiv_https___arxiv_org_abs_2505_14083
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Computational Efficiency under Covariate Shift in Kernel Ridge Regression
Della Vecchia, Andrea
Watusadisi, Arnaud Mavakala
De Vito, Ernesto
Rosasco, Lorenzo
Machine Learning
This paper addresses the covariate shift problem in the context of nonparametric regression within reproducing kernel Hilbert spaces (RKHSs). Covariate shift arises in supervised learning when the input distributions of the training and test data differ, presenting additional challenges for learning. Although kernel methods have optimal statistical properties, their high computational demands in terms of time and, particularly, memory, limit their scalability to large datasets. To address this limitation, the main focus of this paper is to explore the trade-off between computational efficiency and statistical accuracy under covariate shift. We investigate the use of random projections where the hypothesis space consists of a random subspace within a given RKHS. Our results show that, even in the presence of covariate shift, significant computational savings can be achieved without compromising learning performance.
title Computational Efficiency under Covariate Shift in Kernel Ridge Regression
topic Machine Learning
url https://arxiv.org/abs/2505.14083