paper-with-me

Papers

Generalization in Kernel Regression Under Realistic Assumptions

2023-12-26 · Daniel Barzilai, Ohad Shamir

It is by now well-established that modern over-parameterized models seem to elude the bias-variance tradeoff and generalize well despite overfitting noise. Many recent works attempt to analyze this phenomenon in the relatively tractable setting of kernel regression. However, as we argue in detail, most past works on this topic either make unrealistic assumptions, or focus on a narrow problem setup. This work aims to provide a unified theory to upper bound the excess risk of kernel regression for nearly all common and realistic settings. Specifically, we provide rigorous bounds that hold for common kernels and for any amount of regularization, noise, any input dimension, and any number of samples. Furthermore, we provide relative perturbation bounds for the eigenvalues of kernel matrices, which may be of independent interest. These reveal a self-regularization phenomenon, whereby a heavy tail in the eigendecomposition of the kernel provides it with an implicit form of regularization, enabling good generalization. When applied to common kernels, our results imply benign overfitting in high input dimensions, nearly tempered overfitting in fixed dimensions, and explicit convergence rates for regularized regression. As a by-product, we obtain time-dependent bounds for neural networks trained in the kernel regime.

📄 PDF Abstract BibTeX arXiv:2312.15995

Code (0)

등록된 구현이 없습니다.

Tasks

regression

Methods 이 논문이 사용한 방법론

Focus 설명 없음

Similar Papers 제목 키워드 기반

On the Asymptotic Learning Curves of Kernel Ridge Regression under Power-law Decay

2023-09-23 · NeurIPS 2023 11

The widely observed 'benign overfitting phenomenon' in the neural network literature raises the challenge to the 'bias-variance trade-off' doctrine in the statistical learning theory. Since the generalization ability of …

Learning Theoryregression

Generalization Error Curves for Analytic Spectral Algorithms under Power-law Decay

2024-01-03 · Yicheng Li, Weiye Gan, Zuoqiang Shi, Qian Lin

The generalization error curve of certain kernel regression method aims at determining the exact order of generalization error with various source condition, noise level and choice of the regularization parameter rather …

regression

How rotational invariance of common kernels prevents generalization in high dimensions

2021-04-09 · Konstantin Donhauser, Mingqi Wu, Fanny Yang

Kernel ridge regression is well-known to achieve minimax optimal rates in low-dimensional settings. However, its behavior in high dimensions is much less understood. Recent work establishes consistency for kernel regress…

regressionVocal Bursts Intensity Prediction

Learning curves for Gaussian process regression with power-law priors and targets

2021-10-23 · ICLR 2022 4 · Hui Jin, Pradeep Kr. Banerjee, Guido Montúfar

We characterize the power-law asymptotics of learning curves for Gaussian process regression (GPR) under the assumption that the eigenspectrum of the prior and the eigenexpansion coefficients of the target function follo…

GPRregression

The Optimality of Kernel Classifiers in Sobolev Space

2024-02-02 · Jianfa Lai, Zhifan Li, Dongming Huang, Qian Lin

Kernel methods are widely used in machine learning, especially for classification problems. However, the theoretical analysis of kernel classification is still limited. This paper investigates the statistical performance…

Classification