Saved in:
Bibliographic Details
Main Authors: Pesenti, Lucas, Slot, Lucas, Wiedmer, Manuel
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.06027
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917317956337664
author Pesenti, Lucas
Slot, Lucas
Wiedmer, Manuel
author_facet Pesenti, Lucas
Slot, Lucas
Wiedmer, Manuel
contents The complexity of learning a concept class under Gaussian marginals in the difficult agnostic model is closely related to its $L_1$-approximability by low-degree polynomials. For any concept class with Gaussian surface area at most $Γ$, Klivans et al. (2008) show that degree $d = O(Γ^2 / \varepsilon^4)$ suffices to achieve an $\varepsilon$-approximation. This leads to the best-known bounds on the complexity of learning a variety of concept classes. In this note, we improve their analysis by showing that degree $d = \tilde O (Γ^2 / \varepsilon^2)$ is enough. In light of lower bounds due to Diakonikolas et al. (2021), this yields (near) optimal bounds on the complexity of agnostically learning polynomial threshold functions in the statistical query model. Our proof relies on a direct analogue of a construction of Feldman et al. (2020), who considered $L_1$-approximation on the Boolean hypercube.
format Preprint
id arxiv_https___arxiv_org_abs_2603_06027
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Agnostic learning in (almost) optimal time via Gaussian surface area
Pesenti, Lucas
Slot, Lucas
Wiedmer, Manuel
Machine Learning
Data Structures and Algorithms
The complexity of learning a concept class under Gaussian marginals in the difficult agnostic model is closely related to its $L_1$-approximability by low-degree polynomials. For any concept class with Gaussian surface area at most $Γ$, Klivans et al. (2008) show that degree $d = O(Γ^2 / \varepsilon^4)$ suffices to achieve an $\varepsilon$-approximation. This leads to the best-known bounds on the complexity of learning a variety of concept classes. In this note, we improve their analysis by showing that degree $d = \tilde O (Γ^2 / \varepsilon^2)$ is enough. In light of lower bounds due to Diakonikolas et al. (2021), this yields (near) optimal bounds on the complexity of agnostically learning polynomial threshold functions in the statistical query model. Our proof relies on a direct analogue of a construction of Feldman et al. (2020), who considered $L_1$-approximation on the Boolean hypercube.
title Agnostic learning in (almost) optimal time via Gaussian surface area
topic Machine Learning
Data Structures and Algorithms
url https://arxiv.org/abs/2603.06027