---
title: "Dimension-Free Private Mean Estimation is Impossible Without the Covariance"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.06615v1"
permalink: "https://feed.astralor.com/items/itm_725a4eb1c0a3280109f44145"
published: "2026-11-04T15:01:31.000Z"
score: 58
channel: "ai"
tags: ["Dimension-Free Private Mean","Estimation","Impossible Without","Covariance","Gaussian","Prior"]
---

# Dimension-Free Private Mean Estimation is Impossible Without the Covariance

## 速览

We study the limits of differentially private Gaussian mean estimation when the covariance matrix is unknown. Prior work shows that when the covariance $Σ$ is known, the optimal sample complexity can depend only on its …

## 关键点

- We study the limits of differentially private Gaussian mean estimation when the covariance matrix is unknown.
- Prior work shows that when the covariance $Σ$ is known, the optimal sample complexity can depend only on its spectrum, rather than on the ambient dimension $d$.
- We establish new lower bounds for Gaussian mean estimation with Euclidean error which show that private estimators cannot achieve similar guarantees when the covariance is unknown.

## 全文

We study the limits of differentially private Gaussian mean estimation when the covariance matrix is unknown. Prior work shows that when the covariance $Σ$ is known, the optimal sample complexity can depend only on its spectrum, rather than on the ambient dimension $d$. We establish new lower bounds for Gaussian mean estimation with Euclidean error which show that private estimators cannot achieve similar guarantees when the covariance is unknown. First, when the sample size $n$ is smaller than the effective dimension $\mathrm{trace}(Σ)/\|Σ\|_{\text{op}}$, the optimal error when the covariance is unknown can be larger than the optimal error when it is known by a factor of $\sqrt{d/n}$. Second, when the sample size is larger than the effective dimension, even for diagonal covariance matrices with effective dimension bounded by $\log d$, the optimal error when the covariance is unknown can be larger than the optimal error when it is known by a factor of $\sqrt{d}/n^{3/2}$. Together, these results show that in the unknown covariance setting, exploiting anisotropic structure under differential privacy can require a dimension-dependent error or sample complexity.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_725a4eb1c0a3280109f44145