# Training Differentially Private Ad Prediction Models with Semi-Sensitive Features

arXiv: https://arxiv.org/abs/2401.15246 · HTML: https://mlsanity.com/paper/2401.15246/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 40 references that are in the map.
- First posted on arXiv 26 Jan 2024.

## Builds on (most cited first, 11 of 11)

- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Deep Learning with Differential Privacy](https://mlsanity.com/paper/1607.00133/) (2016)
- [Renyi Differential Privacy](https://mlsanity.com/paper/1702.07476/) (2017)
- [Deep Learning with Label Differential Privacy](https://mlsanity.com/paper/2102.06062/) (2021)
- [Antipodes of Label Differential Privacy: PATE and ALIBI](https://mlsanity.com/paper/2106.03408/) (2021)
- [Attribution Modeling Increases Efficiency of Bidding in Display Advertising](https://mlsanity.com/paper/1707.06409/) (2017)
- [Reacting to Variations in Product Demand: An Application for Conversion Rate (CR) Prediction in Sponsored Search](https://mlsanity.com/paper/1806.08211/) (2018)
- [Private Ad Modeling with DP-SGD](https://mlsanity.com/paper/2211.11896/) (2022)
- [Private Learning with Public Features](https://mlsanity.com/paper/2310.15454/) (2023)
- [Classification with Partially Private Features](https://mlsanity.com/paper/2312.07583/) (2023)
- [Masked LARk: Masked Learning, Aggregation and Reporting worKflow](https://mlsanity.com/paper/2110.14794/) (2021)

## Built on by (most cited first, 6 of 6)

- [Locally Private Estimation with Public Features](https://mlsanity.com/paper/2405.13481/) (2024)
- [Enhancing Learning with Label Differential Privacy by Vector Approximation](https://mlsanity.com/paper/2405.15150/) (2024)
- [On Convex Optimization with Semi-Sensitive Features](https://mlsanity.com/paper/2406.19040/) (2024)
- [Cocoon: A System Architecture for Differentially Private Training with Correlated Noises](https://mlsanity.com/paper/2510.07304/) (2025)
- [Integrating Feature Correlation in Differential Privacy with Applications in DP-ERM](https://mlsanity.com/paper/2605.03945/) (2026)
- [Private Learning with Public Feature Conditioning](https://mlsanity.com/paper/2606.18773/) (2026)

## Abstract (arXiv)

Motivated by problems arising in digital advertising, we introduce the task of training differentially private (DP) machine learning models with semi-sensitive features. In this setting, a subset of the features is known to the attacker (and thus need not be protected) while the remaining features as well as the label are unknown to the attacker and should be protected by the DP guarantee. This task interpolates between training the model with full DP (where the label and all features should be protected) or with label DP (where all the features are considered known, and only the label should be protected). We present a new algorithm for training DP models with semi-sensitive features. Through an empirical evaluation on real ads datasets, we demonstrate that our algorithm surpasses in utility the baselines of (i) DP stochastic gradient descent (DP-SGD) run on all features (known and unknown), and (ii) a label DP algorithm run only on the known features (while discarding the unknown ones).