Optimal Rate of Kernel Regression in Large Dimensions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lu, Weihao, Zhang, Haobo, Li, Yicheng, Xu, Manyun, Lin, Qian
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917707886100480
author Lu, Weihao
Zhang, Haobo
Li, Yicheng
Xu, Manyun
Lin, Qian
author_facet Lu, Weihao
Zhang, Haobo
Li, Yicheng
Xu, Manyun
Lin, Qian
contents We perform a study on kernel regression for large-dimensional data (where the sample size $n$ is polynomially depending on the dimension $d$ of the samples, i.e., $n\asymp d^γ$ for some $γ>0$ ). We first build a general tool to characterize the upper bound and the minimax lower bound of kernel regression for large dimensional data through the Mendelson complexity $\varepsilon_{n}^{2}$ and the metric entropy $\bar{\varepsilon}_{n}^{2}$ respectively. When the target function falls into the RKHS associated with a (general) inner product model defined on $\mathbb{S}^{d}$, we utilize the new tool to show that the minimax rate of the excess risk of kernel regression is $n^{-1/2}$ when $n\asymp d^γ$ for $γ=2, 4, 6, 8, \cdots$. We then further determine the optimal rate of the excess risk of kernel regression for all the $γ>0$ and find that the curve of optimal rate varying along $γ$ exhibits several new phenomena including the multiple descent behavior and the periodic plateau behavior. As an application, For the neural tangent kernel (NTK), we also provide a similar explicit description of the curve of optimal rate. As a direct corollary, we know these claims hold for wide neural networks as well.
format Preprint
id arxiv_https___arxiv_org_abs_2309_04268
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Optimal Rate of Kernel Regression in Large Dimensions
Lu, Weihao
Zhang, Haobo
Li, Yicheng
Xu, Manyun
Lin, Qian
Machine Learning
Statistics Theory
62G08, 46E22, 68T07
We perform a study on kernel regression for large-dimensional data (where the sample size $n$ is polynomially depending on the dimension $d$ of the samples, i.e., $n\asymp d^γ$ for some $γ>0$ ). We first build a general tool to characterize the upper bound and the minimax lower bound of kernel regression for large dimensional data through the Mendelson complexity $\varepsilon_{n}^{2}$ and the metric entropy $\bar{\varepsilon}_{n}^{2}$ respectively. When the target function falls into the RKHS associated with a (general) inner product model defined on $\mathbb{S}^{d}$, we utilize the new tool to show that the minimax rate of the excess risk of kernel regression is $n^{-1/2}$ when $n\asymp d^γ$ for $γ=2, 4, 6, 8, \cdots$. We then further determine the optimal rate of the excess risk of kernel regression for all the $γ>0$ and find that the curve of optimal rate varying along $γ$ exhibits several new phenomena including the multiple descent behavior and the periodic plateau behavior. As an application, For the neural tangent kernel (NTK), we also provide a similar explicit description of the curve of optimal rate. As a direct corollary, we know these claims hold for wide neural networks as well.
title Optimal Rate of Kernel Regression in Large Dimensions
topic Machine Learning
Statistics Theory
62G08, 46E22, 68T07
url https://arxiv.org/abs/2309.04268