Optimizing Kernel Discrepancies via Subset Selection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Deyao, Clément, François, Doerr, Carola, Kirk, Nathan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911248655843328
author Chen, Deyao
Clément, François
Doerr, Carola
Kirk, Nathan
author_facet Chen, Deyao
Clément, François
Doerr, Carola
Kirk, Nathan
contents Kernel discrepancies are a powerful tool for analyzing worst-case errors in quasi-Monte Carlo (QMC) methods. Building on recent advances in optimizing such discrepancy measures, we extend the subset selection problem to the setting of kernel discrepancies, selecting an m-element subset from a large population of size $n \gg m$. We introduce a novel subset selection algorithm applicable to general kernel discrepancies to efficiently generate low-discrepancy samples from both the uniform distribution on the unit hypercube, the traditional setting of classical QMC, and from more general distributions $F$ with known density functions by employing the kernel Stein discrepancy. We also explore the relationship between the classical $L_2$ star discrepancy and its $L_\infty$ counterpart.
format Preprint
id arxiv_https___arxiv_org_abs_2511_02706
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Optimizing Kernel Discrepancies via Subset Selection
Chen, Deyao
Clément, François
Doerr, Carola
Kirk, Nathan
Machine Learning
Computational Geometry
Numerical Analysis
Kernel discrepancies are a powerful tool for analyzing worst-case errors in quasi-Monte Carlo (QMC) methods. Building on recent advances in optimizing such discrepancy measures, we extend the subset selection problem to the setting of kernel discrepancies, selecting an m-element subset from a large population of size $n \gg m$. We introduce a novel subset selection algorithm applicable to general kernel discrepancies to efficiently generate low-discrepancy samples from both the uniform distribution on the unit hypercube, the traditional setting of classical QMC, and from more general distributions $F$ with known density functions by employing the kernel Stein discrepancy. We also explore the relationship between the classical $L_2$ star discrepancy and its $L_\infty$ counterpart.
title Optimizing Kernel Discrepancies via Subset Selection
topic Machine Learning
Computational Geometry
Numerical Analysis
url https://arxiv.org/abs/2511.02706