# Assessment of Differentially Private Synthetic Data for Utility and Fairness in End-to-End Machine Learning Pipelines for Tabular Data

arXiv: https://arxiv.org/abs/2310.19250 · HTML: https://mlsanity.com/paper/2310.19250/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 55 references that are in the map.
- First posted on arXiv 30 Oct 2023.

## Builds on (most cited first, 28 of 28)

- [Deep Learning with Differential Privacy](https://mlsanity.com/paper/1607.00133/) (2016)
- [Modeling Tabular data using Conditional GAN](https://mlsanity.com/paper/1907.00503/) (2019)
- [Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data](https://mlsanity.com/paper/1610.05755/) (2016)
- [Differential Privacy Has Disparate Impact on Model Accuracy](https://mlsanity.com/paper/1905.12101/) (2019)
- [Differentially Private Generative Adversarial Network](https://mlsanity.com/paper/1802.06739/) (2018)
- [Fairness in Machine Learning](https://mlsanity.com/paper/2012.15816/) (2020)
- [Fair Regression: Quantitative Definitions and Reduction-based Algorithms](https://mlsanity.com/paper/1905.12843/) (2019)
- [Winning the NIST Contest: A scalable and general approach to differentially private synthetic data](https://mlsanity.com/paper/2108.04978/) (2021)
- [Benchmarking Differentially Private Synthetic Data Generation Algorithms](https://mlsanity.com/paper/2112.09238/) (2021)
- [Graphical-model based estimation and inference for differential privacy](https://mlsanity.com/paper/1901.09136/) (2019)
- [Synthetic Data -- what, why and how?](https://mlsanity.com/paper/2205.03257/) (2022)
- [A Moral Framework for Understanding of Fair ML through Economic Models of Equality of Opportunity](https://mlsanity.com/paper/1809.03400/) (2018)
- [Diffprivlib: The IBM Differential Privacy Library](https://mlsanity.com/paper/1907.02444/) (2019)
- [Robin Hood and Matthew Effects: Differential Privacy Has Disparate Impact on Synthetic Data](https://mlsanity.com/paper/2109.11429/) (2021)
- [Fair Bayesian Optimization](https://mlsanity.com/paper/2006.05109/) (2020)
- [Privacy Loss in Apple's Implementation of Differential Privacy on MacOS 10.12](https://mlsanity.com/paper/1709.02753/) (2017)
- [A Multifaceted Benchmarking of Synthetic Electronic Health Record Generation Models](https://mlsanity.com/paper/2208.01230/) (2022)
- [Differentially Private Synthetic Data: Applied Evaluations and Enhancements](https://mlsanity.com/paper/2011.05537/) (2020)
- [ProPublica's COMPAS Data Revisited](https://mlsanity.com/paper/1906.04711/) (2019)
- [Reducing bias and increasing utility by federated generative modeling of medical images using a centralized adversary](https://mlsanity.com/paper/2101.07235/) (2021)
- [Google COVID-19 Community Mobility Reports: Anonymization Process Description (version 1.1)](https://mlsanity.com/paper/2004.04145/) (2020)
- [Evaluating the Fairness Impact of Differentially Private Synthetic Data](https://mlsanity.com/paper/2205.04321/) (2022)
- [Faking feature importance: A cautionary tale on the use of differentially-private synthetic data](https://mlsanity.com/paper/2203.01363/) (2022)
- [FFPDG: Fast, Fair and Private Data Generation](https://mlsanity.com/paper/2307.00161/) (2023)
- [DP-SGD vs PATE: Which Has Less Disparate Impact on GANs?](https://mlsanity.com/paper/2111.13617/) (2021)
- [An Analysis of the Deployment of Models Trained on Private Tabular Synthetic Data: Unexpected Surprises](https://mlsanity.com/paper/2106.10241/) (2021)
- [U.S. Broadband Coverage Data Set: A Differentially Private Data Release](https://mlsanity.com/paper/2103.14035/) (2021)
- [Towards Auditability for Fairness in Deep Learning](https://mlsanity.com/paper/2012.00106/) (2020)

## Built on by (most cited first, 8 of 8)

- [Privacy Vulnerabilities in Marginals-based Synthetic Data](https://mlsanity.com/paper/2410.05506/) (2024)
- [SynthTextEval: Synthetic Text Data Generation and Evaluation for High-Stakes Domains](https://mlsanity.com/paper/2507.07229/) (2025)
- [Can Synthetic Data be Fair and Private? A Comparative Study of Synthetic Data Generation and Fairness Algorithms](https://mlsanity.com/paper/2501.01785/) (2025)
- [Where to Intervene? Benchmarking Fairness-Aware Learning on Differentially Private Synthetic Tabular Data](https://mlsanity.com/paper/2607.07471/) (2026)
- [SAFES: Sequential Privacy and Fairness Enhancing Data Synthesis for Responsible AI](https://mlsanity.com/paper/2411.09178/) (2024)
- [Beyond Internal Data: Constructing Complete Datasets for Fairness Testing](https://mlsanity.com/paper/2507.18561/) (2025)
- [Evaluation metrics for temporal preservation in synthetic longitudinal patient data](https://mlsanity.com/paper/2602.10643/) (2026)
- [Disparate Impact in Synthetic Data Generation](https://mlsanity.com/paper/2606.13105/) (2026)

## Abstract (arXiv)

Differentially private (DP) synthetic data sets are a solution for sharing data while preserving the privacy of individual data providers. Understanding the effects of utilizing DP synthetic data in end-to-end machine learning pipelines impacts areas such as health care and humanitarian action, where data is scarce and regulated by restrictive privacy laws. In this work, we investigate the extent to which synthetic data can replace real, tabular data in machine learning pipelines and identify the most effective synthetic data generation techniques for training and evaluating machine learning models. We investigate the impacts of differentially private synthetic data on downstream classification tasks from the point of view of utility as well as fairness. Our analysis is comprehensive and includes representatives of the two main types of synthetic data generation algorithms: marginal-based and GAN-based. To the best of our knowledge, our work is the first that: (i) proposes a training and evaluation framework that does not assume that real data is available for testing the utility and fairness of machine learning models trained on synthetic data; (ii) presents the most extensive analysis of synthetic data set generation algorithms in terms of utility and fairness when used for training machine learning models; and (iii) encompasses several different definitions of fairness. Our findings demonstrate that marginal-based synthetic data generators surpass GAN-based ones regarding model training utility for tabular data. Indeed, we show that models trained using data generated by marginal-based algorithms can exhibit similar utility to models trained using real data. Our analysis also reveals that the marginal-based synthetic data generator MWEM PGM can train models that simultaneously achieve utility and fairness characteristics close to those obtained by models trained with real data.