HeteroFedSyn: Differentially Private Tabular Data Synthesis for Heterogeneous Federated Settings
Open Access
- 18 May 2026
- journal article
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Management of Data
- Vol. 4 (3) , 1-26
Abstract
Traditional Differential Privacy (DP) mechanisms are typically tailored to specific analysis tasks, which limits the reusability of protected data. DP tabular data synthesis overcomes this by generating synthetic datasets that can be shared for arbitrary downstream tasks. However, existing synthesis methods predominantly assume centralized or local settings and overlook the more practical horizontal federated scenario. Naïvely synthesizing data locally or perturbing individual records either produces biased mixtures or introduces excessive noise, especially under heterogeneous data distributions across participants. We propose HeteroFedSyn, the first DP tabular data synthesis framework designed specifically for the horizontal federated setting. Built upon the PrivSyn paradigm of 2-way marginal–based synthesis, HeteroFedSyn introduces three key innovations for distributed marginal selection: (i) an l 2 -based dependency metric with random projection for noise-efficient correlation measurement, (ii) an unbiased estimator to correct multiplicative noise, and (iii) an adaptive selection strategy that dynamically updates dependency scores to avoid redundancy. Extensive experiments on range queries, Wasserstein fidelity, and machine learning tasks show that, despite the increased noise inherent to federated execution, HeteroFedSyn achieves utility comparable to centralized synthesis. Our code is open-sourced via the link.Keywords
All Related Versions
Funding Information
- National Science Foundation (AST-2132700)
- National Science Foundation (ECCS-2332060)
- National Science Foundation (CNS-2313110)
- National Science Foundation (ECCS-2143559)
- National Science Foundation (ECCS-2531023)
- National Science Foundation (CNS-2531789)
- National Science Foundation (CNS-2220433)
This publication has 21 references indexed in Scilit:
- SoK: Privacy-Preserving Data Synthesis2024
- DPI: Ensuring Strict Differential Privacy for Infinite Data Streaming2024
- Continual Observation under User-level Differential Privacy2023
- DP-CTGAN: Differentially Private Medical Data Generation Using CTGANs2022
- Medical imaging deep learning with differential privacyScientific Reports, 2021
- Scalable Private Set Union from Symmetric-Key Techniques2019
- DP-CGAN: Differentially Private Synthetic Data and Label Generation2019
- RAPPOR2014
- Differential privacy under continual observation2010
- Extensions of Lipschitz mappings into a Hilbert space1984