mlsanityData to 7 Oct 2026

Bounding the Excess Risk for Linear Models Trained on Marginal-Preserving, Differentially-Private, Synthetic Data

arXiv 2402.04375 ↗First posted 6 Feb 2024cs.LG cs.CRcited by 3

Not on the map. Why?

Abstract (arXiv)

The growing use of machine learning (ML) has raised concerns that an ML model may reveal private information about an individual who has contributed to the training dataset. To prevent leakage of sensitive data, we consider using differentially-private (DP), synthetic training data instead of real training data to train an ML model. A key desirable property of synthetic data is its ability to preserve the low-order marginals of the original distribution. Our main contribution comprises novel upper and lower bounds on the excess empirical risk of linear models trained on such synthetic data, for continuous and Lipschitz loss functions. We perform extensive experimentation alongside our theoretical results.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 22 of its 50 references that are in the map. First posted on arXiv 6 Feb 2024.

Builds on

22 of 50 references.

References · 22

  1. 2016Deep Learning with Differential Privacycited by 2,682
  2. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  3. 2016Semi-supervised Knowledge Transfer for Deep Learning from Private Training Datacited by 347
  4. 2017Real-valued (Medical) Time Series Generation with Recurrent Conditional GANscited by 329
  5. 2018Scalable Private Learning with PATEcited by 266
  6. 2022Unlocking High-Accuracy Differentially Private Image Classification through Scalecited by 218
  7. 2020The Discrete Gaussian for Differential Privacycited by 134
  8. 2021Hyperparameter Tuning with Renyi Differential Privacycited by 104
  9. 2021Winning the NIST Contest: A scalable and general approach to differentially private synthetic datacited by 101
  10. 2022AIM: An Adaptive and Iterative Mechanism for Differentially Private Synthetic Datacited by 88
  11. 2022Variational Model Inversion Attackscited by 75
  12. 2019Graphical-model based estimation and inference for differential privacycited by 73
10 more
  1. 2017Efficient Private ERM for Smooth Objectivescited by 61
  2. 2021Iterative Methods for Private Synthetic Data: Unifying Framework and New Methodscited by 58
  3. 2021Differentially private inference via noisy optimizationcited by 27
  4. 2021An Efficient DP-SGD Mechanism for Large Scale NLP Modelscited by 27
  5. 2020Dual Policy Distillationcited by 25
  6. 2021Dopamine: Differentially Private Federated Learning on Medical Datacited by 18
  7. 2023Faster Differentially Private Convex Optimization via Second-Order Methodscited by 13
  8. 2020How to Democratise and Protect AI: Fair and Differentially Private Decentralised Deep Learningcited by 12
  9. 2019Interaction is necessary for distributed learning with privacy or communication constraintscited by 8
  10. 2023Statistical Theory of Differentially Private Marginal-based Data Synthesis Algorithmscited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 0; 2026: 1.

3 papers.

  1. 2024Private Regression via Data-Dependent Sufficient Statistic Perturbationcited by 4
  2. 2024Balancing Fairness and Accuracy in Data-Restricted Binary Classificationcited by 0
  3. 2026Should I use Synthetic Data for That? An Analysis of the Suitability of Synthetic Data for Data Sharing and Augmentationcited by 0