Thinning to Improve Two-Sample Discrepancy

Smirnov, Gleb

doi:10.1002/rsa.70049

Thinning to Improve Two-Sample Discrepancy

2026-03-01

0

PRE

AI

DOI：10.1002/rsa.70049

Original

Original request for help

Save

Abstract

En 中文

The discrepancy between two independent samples X-1, & mldr;, X-n and Y-1, & mldr;, Y-n drawn from the same distribution on & Ropf;(d) typically has order O( root n) even in one dimension. We give a simple online algorithm that reduces the discrepancy to O(log(2d)n) by discarding a small fraction of the points.

Thinning to Improve Two-Sample Discrepancy

Abstract

Journal

Organization