mlsanityData to 7 Oct 2026

Private Ad Modeling with DP-SGD

arXiv 2211.11896 ↗First posted 21 Nov 2022cs.LG cs.CRcited by 6

Not on the map. Why?

Abstract (arXiv)

A well-known algorithm in privacy-preserving ML is differentially private stochastic gradient descent (DP-SGD). While this algorithm has been evaluated on text and image data, it has not been previously applied to ads data, which are notorious for their high class imbalance and sparse gradient updates. In this work we apply DP-SGD to several ad modeling tasks including predicting click-through rates, conversion rates, and number of conversion events, and evaluate their privacy-utility trade-off on real-world datasets. Our work is the first to empirically demonstrate that DP-SGD can provide both privacy and utility for ad modeling tasks.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 22 of its 53 references that are in the map. First posted on arXiv 21 Nov 2022.

Builds on

22 of 53 references.

References · 22

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  3. 2016Deep Learning with Differential Privacycited by 2,682
  4. 2021Large Language Models Can Be Strong Differentially Private Learnerscited by 405
  5. 2016Semi-supervised Knowledge Transfer for Deep Learning from Private Training Datacited by 347
  6. 2018Scalable Private Learning with PATEcited by 266
  7. 2019Differentially Private Learning with Adaptive Clippingcited by 218
  8. 2022Unlocking High-Accuracy Differentially Private Image Classification through Scalecited by 218
  9. 2021Numerical Composition of Differential Privacycited by 174
  10. 2021Practical and Private (Deep) Learning without Sampling or Shufflingcited by 169
  11. 2021Large-Scale Differentially Private BERTcited by 117
  12. 2021Deep Learning with Label Differential Privacycited by 113
10 more
  1. 2022Toward Training at ImageNet Scale with Differential Privacycited by 87
  2. 2019AdaCliP: Adaptive Clipping for Private SGDcited by 77
  3. 2020Enabling Fast Differentially Private SGD via Just-in-Time Compilation and Vectorizationcited by 58
  4. 2021Antipodes of Label Differential Privacy: PATE and ALIBIcited by 50
  5. 2022Connect the Dots: Tighter Discrete Approximations of Privacy Loss Distributionscited by 50
  6. 2022Scalable and Efficient Training of Large Convolutional Neural Networks with Differential Privacycited by 47
  7. 2020Scaling up Differentially Private Deep Learning with Fast Per-Example Gradient Clippingcited by 46
  8. 2015Efficient Per-Example Gradient Computationscited by 42
  9. 2021Fast and Memory Efficient Differentially Private-SGD via JL Projectionscited by 29
  10. 2022Faster Privacy Accounting via Evolving Discretizationcited by 14

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 3; 2025: 1.

6 papers.

  1. 2024How Private are DP-SGD Implementations?cited by 22
  2. 2024Training Differentially Private Ad Prediction Models with Semi-Sensitive Featurescited by 6
  3. 2023Sparsity-Preserving Differentially Private Training of Large Embedding Modelscited by 5
  4. 2023Epsilon*: Privacy Metric for Machine Learning Modelscited by 2
  5. 2024LazyDP: Co-Designing Algorithm-Software for Scalable Training of Differentially Private Recommendation Modelscited by 1
  6. 2025Cocoon: A System Architecture for Differentially Private Training with Correlated Noisescited by 1