# Element Level Differential Privacy: The Right Granularity of Privacy

arXiv: https://arxiv.org/abs/1912.04042 · HTML: https://mlsanity.com/paper/1912.04042/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Dec 2019.

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

- [Shuffle Private Stochastic Convex Optimization](https://mlsanity.com/paper/2106.09805/) (2021)
- [Algorithms with More Granular Differential Privacy Guarantees](https://mlsanity.com/paper/2209.04053/) (2022)
- [How Do Input Attributes Impact the Privacy Loss in Differential Privacy?](https://mlsanity.com/paper/2211.10173/) (2022)
- [User-Entity Differential Privacy in Learning Natural Language Models](https://mlsanity.com/paper/2211.01141/) (2022)
- [On Using Secure Aggregation in Differentially Private Federated Learning with Multiple Local Steps](https://mlsanity.com/paper/2407.19286/) (2024)
- [HydraGAN A Multi-head, Multi-objective Approach to Synthetic Data Generation](https://mlsanity.com/paper/2111.07015/) (2021)

## Abstract (arXiv)

Differential Privacy (DP) provides strong guarantees on the risk of compromising a user's data in statistical learning applications, though these strong protections make learning challenging and may be too stringent for some use cases. To address this, we propose element level differential privacy, which extends differential privacy to provide protection against leaking information about any particular "element" a user has, allowing better utility and more robust results than classical DP. By carefully choosing these "elements," it is possible to provide privacy protections at a desired granularity. We provide definitions, associated privacy guarantees, and analysis to identify the tradeoffs with the new definition; we also develop several private estimation and learning methodologies, providing careful examples for item frequency and M-estimation (empirical risk minimization) with concomitant privacy and utility analysis. We complement our theoretical and methodological advances with several real-world applications, estimating histograms and fitting several large-scale prediction models, including deep networks.