Tight Bounds for Learning Polyhedra with a Margin

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Patel, Shyamal, Vempala, Santosh
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917412847222784
author Patel, Shyamal
Vempala, Santosh
author_facet Patel, Shyamal
Vempala, Santosh
contents We give an algorithm for PAC learning intersections of $k$ halfspaces with a $ρ$ margin to within error $\varepsilon$ that runs in time $\textsf{poly}(k, \varepsilon^{-1}, ρ^{-1}) \cdot \exp \left(O(\sqrt{n \log(1/ρ) \log k})\right)$. Notably, this improves on prior work which had an exponential dependence on either $k$ or $ρ^{-1}$ and matches known cryptographic and Statistical Query lower bounds up to the logarithmic factors in $k$ and $ρ$ in the exponent. Our learning algorithm extends to the more general setting when we are only promised that most points have distance at least $ρ$ from the boundary of the polyhedron, making it applicable to continuous distributions as well.
format Preprint
id arxiv_https___arxiv_org_abs_2604_14614
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Tight Bounds for Learning Polyhedra with a Margin
Patel, Shyamal
Vempala, Santosh
Data Structures and Algorithms
Machine Learning
We give an algorithm for PAC learning intersections of $k$ halfspaces with a $ρ$ margin to within error $\varepsilon$ that runs in time $\textsf{poly}(k, \varepsilon^{-1}, ρ^{-1}) \cdot \exp \left(O(\sqrt{n \log(1/ρ) \log k})\right)$. Notably, this improves on prior work which had an exponential dependence on either $k$ or $ρ^{-1}$ and matches known cryptographic and Statistical Query lower bounds up to the logarithmic factors in $k$ and $ρ$ in the exponent. Our learning algorithm extends to the more general setting when we are only promised that most points have distance at least $ρ$ from the boundary of the polyhedron, making it applicable to continuous distributions as well.
title Tight Bounds for Learning Polyhedra with a Margin
topic Data Structures and Algorithms
Machine Learning
url https://arxiv.org/abs/2604.14614