Saved in:
Bibliographic Details
Main Authors: Iyer, Gautam, Venkatraman, Raghavendra
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.17794
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908723250724864
author Iyer, Gautam
Venkatraman, Raghavendra
author_facet Iyer, Gautam
Venkatraman, Raghavendra
contents Given $N$ i.i.d. samples from a probability measure $μ$ on $\mathbf{R}^d$, we study the rate of convergence of the empirical measure $μ_N \to μ$ in the negative Sobolev space $W^{-α, p}$. When $W^{-α, p}$ contains point measures (i.e. when $αp > (p-1)d$), we show $\mathbf{E} \| μ_N - μ\|_{W^{-α, p}}^p \leq C_d / N^{p/2}$ for an explicit dimensional constant $C_d$, and obtain a Gaussian tail bound. When $0 < αp \leq d(p-1)$, we prove a similar result for Gaussian regularizations.
format Preprint
id arxiv_https___arxiv_org_abs_2512_17794
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Convergence of Empirical Measures for i.i.d. samples in $W^{-α, p}$
Iyer, Gautam
Venkatraman, Raghavendra
Probability
60B10 (Primary) 60G50, 46E35 (Secondary)
Given $N$ i.i.d. samples from a probability measure $μ$ on $\mathbf{R}^d$, we study the rate of convergence of the empirical measure $μ_N \to μ$ in the negative Sobolev space $W^{-α, p}$. When $W^{-α, p}$ contains point measures (i.e. when $αp > (p-1)d$), we show $\mathbf{E} \| μ_N - μ\|_{W^{-α, p}}^p \leq C_d / N^{p/2}$ for an explicit dimensional constant $C_d$, and obtain a Gaussian tail bound. When $0 < αp \leq d(p-1)$, we prove a similar result for Gaussian regularizations.
title Convergence of Empirical Measures for i.i.d. samples in $W^{-α, p}$
topic Probability
60B10 (Primary) 60G50, 46E35 (Secondary)
url https://arxiv.org/abs/2512.17794