Dropout-Robust Mechanisms for Differentially Private and Fully Decentralized Mean Estimation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sabater, César, Mokhtar, Sonia Ben, Ramon, Jan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912412946399232
author Sabater, César
Mokhtar, Sonia Ben
Ramon, Jan
author_facet Sabater, César
Mokhtar, Sonia Ben
Ramon, Jan
contents Achieving differentially private computations in decentralized settings poses significant challenges, particularly regarding accuracy, communication cost, and robustness against information leakage. While cryptographic solutions offer promise, they often suffer from high communication overhead or require centralization in the presence of network failures. Conversely, existing fully decentralized approaches typically rely on relaxed adversarial models or pairwise noise cancellation, the latter suffering from substantial accuracy degradation if parties unexpectedly disconnect. In this work, we propose IncA, a new protocol for fully decentralized mean estimation, a widely used primitive in data-intensive processing. Our protocol, which enforces differential privacy, requires no central orchestration and employs low-variance correlated noise, achieved by incrementally injecting sensitive information into the computation. First, we theoretically demonstrate that, when no parties permanently disconnect, our protocol achieves accuracy comparable to that of a centralized setting-already an improvement over most existing decentralized differentially private techniques. Second, we empirically show that our use of low-variance correlated noise significantly mitigates the accuracy loss experienced by existing techniques in the presence of dropouts.
format Preprint
id arxiv_https___arxiv_org_abs_2506_03746
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dropout-Robust Mechanisms for Differentially Private and Fully Decentralized Mean Estimation
Sabater, César
Mokhtar, Sonia Ben
Ramon, Jan
Cryptography and Security
Distributed, Parallel, and Cluster Computing
Machine Learning
Achieving differentially private computations in decentralized settings poses significant challenges, particularly regarding accuracy, communication cost, and robustness against information leakage. While cryptographic solutions offer promise, they often suffer from high communication overhead or require centralization in the presence of network failures. Conversely, existing fully decentralized approaches typically rely on relaxed adversarial models or pairwise noise cancellation, the latter suffering from substantial accuracy degradation if parties unexpectedly disconnect. In this work, we propose IncA, a new protocol for fully decentralized mean estimation, a widely used primitive in data-intensive processing. Our protocol, which enforces differential privacy, requires no central orchestration and employs low-variance correlated noise, achieved by incrementally injecting sensitive information into the computation. First, we theoretically demonstrate that, when no parties permanently disconnect, our protocol achieves accuracy comparable to that of a centralized setting-already an improvement over most existing decentralized differentially private techniques. Second, we empirically show that our use of low-variance correlated noise significantly mitigates the accuracy loss experienced by existing techniques in the presence of dropouts.
title Dropout-Robust Mechanisms for Differentially Private and Fully Decentralized Mean Estimation
topic Cryptography and Security
Distributed, Parallel, and Cluster Computing
Machine Learning
url https://arxiv.org/abs/2506.03746