mlsanityData to 14 Oct 2026

A Simple and Practical Method for Reducing the Disparate Impact of Differential Privacy

arXiv 2312.11712 ↗First posted 18 Dec 2023cs.CR cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Differentially private (DP) mechanisms have been deployed in a variety of high-impact social settings (perhaps most notably by the U.S. Census). Since all DP mechanisms involve adding noise to results of statistical queries, they are expected to impact our ability to accurately analyze and learn from data, in effect trading off privacy with utility. Alarmingly, the impact of DP on utility can vary significantly among different sub-populations. A simple way to reduce this disparity is with stratification. First compute an independent private estimate for each group in the data set (which may be the intersection of several protected classes), then, to compute estimates of global statistics, appropriately recombine these group estimates. Our main observation is that naive stratification often yields high-accuracy estimates of population-level statistics, without the need for additional privacy budget. We support this observation theoretically and empirically. Our theoretical results center on the private mean estimation problem, while our empirical results center on extensive experiments on private data synthesis to demonstrate the effectiveness of stratification on a variety of private mechanisms. Overall, we argue that this straightforward approach provides a strong baseline against which future work on reducing utility disparities of DP mechanisms should be compared.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 30 of its 62 references that are in the map. First posted on arXiv 18 Dec 2023.

Builds on

30 of 62 references.

References · 30

  1. 2016Equality of Opportunity in Supervised Learningcited by 2,007
  2. 2014Certifying and removing disparate impactcited by 715
  3. 2021Retiring Adult: New Datasets for Fair Machine Learningcited by 449
  4. 2016Concentrated Differential Privacy: Simplifications, Extensions, and Lower Boundscited by 346
  5. 2019Differential Privacy Has Disparate Impact on Model Accuracycited by 316
  6. 2014RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Responsecited by 272
  7. 2017Collecting Telemetry Data Privatelycited by 149
  8. 2017Finite Sample Differentially Private Confidence Intervalscited by 104
  9. 2021Winning the NIST Contest: A scalable and general approach to differentially private synthetic datacited by 102
  10. 2022AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Datacited by 90
  11. 2021Benchmarking Differentially Private Synthetic Data Generation Algorithmscited by 81
  12. 2019Graphical-model based estimation and inference for differential privacycited by 74
18 more
  1. 2020CoinPress: Practical Private Mean and Covariance Estimationcited by 74
  2. 2020Private Mean Estimation of Heavy-Tailed Distributionscited by 73
  3. 2014Differential Privacy and Machine Learning: a Survey and Reviewcited by 67
  4. 2021Robust and Differentially Private Mean Estimationcited by 66
  5. 2021Iterative Methods for Private Synthetic Data: Unifying Framework and New Methodscited by 59
  6. 2019Fair Decision Making using Privacy-Protected Datacited by 58
  7. 2021Differentially Private Query Release Through Adaptive Projectioncited by 52
  8. 2021Robin Hood and Matthew Effects: Differential Privacy Has Disparate Impact on Synthetic Datacited by 48
  9. 2022Differential Privacy and Fairness in Decisions and Learning Tasks: A Surveycited by 47
  10. 2020New Oracle-Efficient Algorithms for Private Synthetic Data Releasecited by 43
  11. 2021Private and polynomial time algorithms for learning Gaussians and beyondcited by 39
  12. 2020Differentially Private Synthetic Data: Applied Evaluations and Enhancementscited by 36
  13. 2021A Private and Computationally-Efficient Estimator for Unbounded Gaussianscited by 34
  14. 2021DP-SGD vs PATE: Which Has Less Disparate Impact on Model Accuracy?cited by 33
  15. 2019Differentially Private Algorithms for Learning Mixtures of Separated Gaussianscited by 27
  16. 2021The Impact of the U.S. Census Disclosure Avoidance System on Redistricting and Voting Rights Analysiscited by 19
  17. 2021Private sampling: a noiseless approach for generating differentially private synthetic datacited by 8
  18. 2022Epistemic Parity: Reproducibility as an Evaluation Metric for Differential Privacycited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 0; 2026: 1.

4 papers.

  1. 2024Differentially Private Fair Binary Classificationscited by 3
  2. 2024PFGuard: A Generative Framework with Privacy and Fairness Safeguardscited by 3
  3. 2024Differentially Private Random Feature Modelcited by 2
  4. 2026Rethinking Anonymity Claims in Synthetic Data Generation: A Model-Centric Privacy Attack Perspectivecited by 1