mlsanityData to 13 Oct 2026

The Trade-Offs of Private Prediction

arXiv 2007.05089 ↗First posted 9 Jul 2020cs.LG stat.MLcited by 22

Not on the map. Why?

Abstract (arXiv)

Machine learning models leak information about their training data every time they reveal a prediction. This is problematic when the training data needs to remain private. Private prediction methods limit how much information about the training data is leaked by each prediction. Private prediction can also be achieved using models that are trained by private training methods. In private prediction, both private training and private prediction methods exhibit trade-offs between privacy, privacy failure probability, amount of training data, and inference budget. Although these trade-offs are theoretically well-understood, they have hardly been studied empirically. This paper presents the first empirical study into the trade-offs of private prediction. Our study sheds light on which methods are best suited for which learning setting. Perhaps surprisingly, we find private training methods outperform private prediction methods in a wide range of private prediction settings.

Key facts. Cited by 22 papers in the map (data through 13 Oct 2026). First posted on arXiv 9 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2021: 2; 2022: 3; 2023: 5; 2024: 4; 2025: 5; 2026: 3.

22 papers.

Sort:

By area:

  1. 2022Unlocking High-Accuracy Differentially Private Image Classification through Scalecited by 223
  2. 2023How to DP-fy ML: A Practical Guide to Machine Learning with Differential Privacycited by 151
  3. 2023On Provable Copyright Protection for Generative Modelscited by 97
  4. 2023Harnessing large-language models to generate private synthetic textcited by 65
  5. 2022Submix: Practical Private Prediction for Large-Scale Language Modelscited by 31
  6. 2024Differentially Private Next-Token Prediction of Large Language Modelscited by 22
  7. 2022Optimal Membership Inference Bounds for Adaptive Composition of Sampled Gaussian Mechanismscited by 19
  8. 2023Unlocking Accuracy and Fairness in Differentially Private Image Classificationcited by 16
  9. 2024Certification for Differentially Private Prediction in Gradient-Based Trainingcited by 7
  10. 2025Differential privacy for medical deep learning: methods, tradeoffs, and deployment implicationscited by 7
  11. 2021Privacy Assessment of Federated Learning using Private Personalized Layerscited by 5
  12. 2021MixNN: Protection of Federated Learning Against Inference Attacks by Mixing Neural Network Layerscited by 5
10 more
  1. 2023"Private Prediction Strikes Back!'' Private Kernelized Nearest Neighbors with Individual Renyi Filtercited by 5
  2. 2024Adaptively Private Next-Token Prediction of Large Language Modelscited by 4
  3. 2025KL-regularization Itself is Differentially Private in Bandits and RLHFcited by 3
  4. 2024Private Truly-Everlasting Robust-Predictioncited by 2
  5. 2025Mitigating Disparate Impact of Differentially Private Learning through Bounded Adaptive Clippingcited by 2
  6. 2025Privacy Preserving In-Context-Learning Framework for Large Language Modelscited by 2
  7. 2025Cocoon: A System Architecture for Differentially Private Training with Correlated Noisescited by 1
  8. 2026Private Prediction via PAC Privacycited by 1
  9. 2026PAC-Private Autoregressive Generation: Calibrating Noise to Ensemble Disagreementcited by 1
  10. 2026Certification-Based Differentially Private Learningcited by 1