Sample and Computationally Efficient Robust Learning of Gaussian Single-Index Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Puqian, Zarifis, Nikos, Diakonikolas, Ilias, Diakonikolas, Jelena
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910690189508608
author Wang, Puqian
Zarifis, Nikos
Diakonikolas, Ilias
Diakonikolas, Jelena
author_facet Wang, Puqian
Zarifis, Nikos
Diakonikolas, Ilias
Diakonikolas, Jelena
contents A single-index model (SIM) is a function of the form $σ(\mathbf{w}^{\ast} \cdot \mathbf{x})$, where $σ: \mathbb{R} \to \mathbb{R}$ is a known link function and $\mathbf{w}^{\ast}$ is a hidden unit vector. We study the task of learning SIMs in the agnostic (a.k.a. adversarial label noise) model with respect to the $L^2_2$-loss under the Gaussian distribution. Our main result is a sample and computationally efficient agnostic proper learner that attains $L^2_2$-error of $O(\mathrm{OPT})+ε$, where $\mathrm{OPT}$ is the optimal loss. The sample complexity of our algorithm is $\tilde{O}(d^{\lceil k^{\ast}/2\rceil}+d/ε)$, where $k^{\ast}$ is the information-exponent of $σ$ corresponding to the degree of its first non-zero Hermite coefficient. This sample bound nearly matches known CSQ lower bounds, even in the realizable setting. Prior algorithmic work in this setting had focused on learning in the realizable case or in the presence of semi-random noise. Prior computationally efficient robust learners required significantly stronger assumptions on the link function.
format Preprint
id arxiv_https___arxiv_org_abs_2411_05708
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Sample and Computationally Efficient Robust Learning of Gaussian Single-Index Models
Wang, Puqian
Zarifis, Nikos
Diakonikolas, Ilias
Diakonikolas, Jelena
Machine Learning
A single-index model (SIM) is a function of the form $σ(\mathbf{w}^{\ast} \cdot \mathbf{x})$, where $σ: \mathbb{R} \to \mathbb{R}$ is a known link function and $\mathbf{w}^{\ast}$ is a hidden unit vector. We study the task of learning SIMs in the agnostic (a.k.a. adversarial label noise) model with respect to the $L^2_2$-loss under the Gaussian distribution. Our main result is a sample and computationally efficient agnostic proper learner that attains $L^2_2$-error of $O(\mathrm{OPT})+ε$, where $\mathrm{OPT}$ is the optimal loss. The sample complexity of our algorithm is $\tilde{O}(d^{\lceil k^{\ast}/2\rceil}+d/ε)$, where $k^{\ast}$ is the information-exponent of $σ$ corresponding to the degree of its first non-zero Hermite coefficient. This sample bound nearly matches known CSQ lower bounds, even in the realizable setting. Prior algorithmic work in this setting had focused on learning in the realizable case or in the presence of semi-random noise. Prior computationally efficient robust learners required significantly stronger assumptions on the link function.
title Sample and Computationally Efficient Robust Learning of Gaussian Single-Index Models
topic Machine Learning
url https://arxiv.org/abs/2411.05708