Continual Release Moment Estimation with Differential Privacy

Nikita P. Kalinin, Jalaj Upadhyay, Christoph H. Lampert
2/10/2025
cs.LGstat.ML

No confident code implementation found for this paper.

Abstract

We propose Joint Moment Estimation (JME), a method for continually and privately estimating both the first and second moments of data with reduced noise compared to naive approaches. JME uses the matrix mechanism and a joint sensitivity analysis to allow the second moment estimation with no additional privacy cost, thereby improving accuracy while maintaining privacy. We demonstrate JME's effectiveness in two applications: estimating the running mean and covariance matrix for Gaussian density estimation, and model training with DP-Adam on CIFAR-10.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Differentially Private Truncation of Unbounded Data via Public Second Moments

82% match2026

Matrix Factorization for Practical Continual Mean Estimation Under User-Level Differential Privacy

81% match2026

Private Mean Estimation with Person-Level Differential Privacy

80% match2024

Optimal Variance and Covariance Estimation under Differential Privacy in the Add-Remove Model and Beyond

78% match2025

Correlated Privacy Mechanisms for Differentially Private Distributed Mean Estimation

78% match2024

A Smooth Binary Mechanism for Efficient Private Continual Observation

78% match1 repo32023

Cite this paper

@article{kalinin2025continual,
  title  = {Continual Release Moment Estimation with Differential Privacy},
  author = {Nikita P. Kalinin and Jalaj Upadhyay and Christoph H. Lampert},
  year   = {2025},
  eprint = {2502.06597},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2502.06597v2},
  journal = {NEURIPS 2025 2025}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2502.06597.svg)](https://codeofpaper.com/paper/2502.06597)

Discussion