Moment Expansions of the Energy Distance

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Langmore, Ian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916760367661056
author Langmore, Ian
author_facet Langmore, Ian
contents The energy distance is used to test distributional equality, and as a loss function in machine learning. While $D^2(X, Y)=0$ only when $X\sim Y$, the sensitivity to different moments is of practical importance. This work considers $D^2(X, Y)$ in the case where the distributions are close. In this regime, $D^2(X, Y)$ is more sensitive to differences in the means $\bar{X}-\bar{Y}$, than differences in the covariances $Δ$. This is due to the structure of the energy distance and is independent of dimension. The sensitivity to on versus off diagonal components of $Δ$ is examined when $X$ and $Y$ are close to isotropic. Here a dimension dependent averaging occurs and, in many cases, off diagonal correlations contribute significantly less. Numerical results verify these relationships hold even when distributional assumptions are not strictly met.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20647
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Moment Expansions of the Energy Distance
Langmore, Ian
Machine Learning
Statistics Theory
62G10 (Primary), 62E20, 62H15, 60E10 (Secondary)
G.3
The energy distance is used to test distributional equality, and as a loss function in machine learning. While $D^2(X, Y)=0$ only when $X\sim Y$, the sensitivity to different moments is of practical importance. This work considers $D^2(X, Y)$ in the case where the distributions are close. In this regime, $D^2(X, Y)$ is more sensitive to differences in the means $\bar{X}-\bar{Y}$, than differences in the covariances $Δ$. This is due to the structure of the energy distance and is independent of dimension. The sensitivity to on versus off diagonal components of $Δ$ is examined when $X$ and $Y$ are close to isotropic. Here a dimension dependent averaging occurs and, in many cases, off diagonal correlations contribute significantly less. Numerical results verify these relationships hold even when distributional assumptions are not strictly met.
title Moment Expansions of the Energy Distance
topic Machine Learning
Statistics Theory
62G10 (Primary), 62E20, 62H15, 60E10 (Secondary)
G.3
url https://arxiv.org/abs/2505.20647