A Comprehensive Analysis on the Learning Curve in Kernel Ridge Regression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cheng, Tin Sum, Lucchi, Aurelien, Kratsios, Anastasis, Belius, David
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914985764978688
author Cheng, Tin Sum
Lucchi, Aurelien
Kratsios, Anastasis
Belius, David
author_facet Cheng, Tin Sum
Lucchi, Aurelien
Kratsios, Anastasis
Belius, David
contents This paper conducts a comprehensive study of the learning curves of kernel ridge regression (KRR) under minimal assumptions. Our contributions are three-fold: 1) we analyze the role of key properties of the kernel, such as its spectral eigen-decay, the characteristics of the eigenfunctions, and the smoothness of the kernel; 2) we demonstrate the validity of the Gaussian Equivalent Property (GEP), which states that the generalization performance of KRR remains the same when the whitened features are replaced by standard Gaussian vectors, thereby shedding light on the success of previous analyzes under the Gaussian Design Assumption; 3) we derive novel bounds that improve over existing bounds across a broad range of setting such as (in)dependent feature vectors and various combinations of eigen-decay rates in the over/underparameterized regimes.
format Preprint
id arxiv_https___arxiv_org_abs_2410_17796
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Comprehensive Analysis on the Learning Curve in Kernel Ridge Regression
Cheng, Tin Sum
Lucchi, Aurelien
Kratsios, Anastasis
Belius, David
Machine Learning
This paper conducts a comprehensive study of the learning curves of kernel ridge regression (KRR) under minimal assumptions. Our contributions are three-fold: 1) we analyze the role of key properties of the kernel, such as its spectral eigen-decay, the characteristics of the eigenfunctions, and the smoothness of the kernel; 2) we demonstrate the validity of the Gaussian Equivalent Property (GEP), which states that the generalization performance of KRR remains the same when the whitened features are replaced by standard Gaussian vectors, thereby shedding light on the success of previous analyzes under the Gaussian Design Assumption; 3) we derive novel bounds that improve over existing bounds across a broad range of setting such as (in)dependent feature vectors and various combinations of eigen-decay rates in the over/underparameterized regimes.
title A Comprehensive Analysis on the Learning Curve in Kernel Ridge Regression
topic Machine Learning
url https://arxiv.org/abs/2410.17796