# Summary Statistic Privacy in Data Sharing

arXiv: https://arxiv.org/abs/2303.02014 · HTML: https://mlsanity.com/paper/2303.02014/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 79 references that are in the map.
- First posted on arXiv 3 Mar 2023.

## Builds on (most cited first, 23 of 23)

- [Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs](https://mlsanity.com/paper/1706.02633/) (2017)
- [Hacking Smart Machines with Smarter Ones: How to Extract Meaningful Data from Machine Learning Classifiers](https://mlsanity.com/paper/1306.4447/) (2013)
- [PacGAN: The power of two samples in generative adversarial networks](https://mlsanity.com/paper/1712.04086/) (2017)
- [Property Inference From Poisoning](https://mlsanity.com/paper/2101.11073/) (2021)
- [Leakage of Dataset Properties in Multi-Party Machine Learning](https://mlsanity.com/paper/2006.07267/) (2020)
- [Formalizing and Estimating Distribution Inference Risks](https://mlsanity.com/paper/2109.06024/) (2021)
- [SNAP: Efficient Extraction of Private Properties with Poisoning](https://mlsanity.com/paper/2208.12348/) (2022)
- [Attribute Privacy: Framework and Mechanisms](https://mlsanity.com/paper/2009.04013/) (2020)
- [Dissecting Distribution Inference](https://mlsanity.com/paper/2212.07591/) (2022)
- [Local Differential Privacy: a tutorial](https://mlsanity.com/paper/1907.11908/) (2019)
- [Inferential Privacy Guarantees for Differentially Private Mechanisms](https://mlsanity.com/paper/1603.01508/) (2016)
- [Protecting Global Properties of Datasets with Distribution Privacy Mechanisms](https://mlsanity.com/paper/2207.08367/) (2022)
- [Information Extraction Under Privacy Constraints](https://mlsanity.com/paper/1511.02381/) (2015)
- [Local Obfuscation Mechanisms for Hiding Probability Distributions](https://mlsanity.com/paper/1812.00939/) (2018)
- [Estimation Efficiency Under Privacy Constraints](https://mlsanity.com/paper/1707.02409/) (2017)
- [Privacy with Estimation Guarantees](https://mlsanity.com/paper/1710.00447/) (2017)
- [PRIVIC: A privacy-preserving method for incremental collection of location data](https://mlsanity.com/paper/2206.10525/) (2022)
- [Back to the Drawing Board: Revisiting the Design of Optimal Location Privacy-preserving Mechanisms](https://mlsanity.com/paper/1705.08779/) (2017)
- [Private Synthetic Data with Hierarchical Structure](https://mlsanity.com/paper/2206.05942/) (2022)
- [Privacy-Aware Guessing Efficiency](https://mlsanity.com/paper/1704.03606/) (2017)
- [Development and Analysis of Deterministic Privacy-Preserving Policies Using Non-Stochastic Information Theory](https://mlsanity.com/paper/1810.11153/) (2018)
- [Noiseless Privacy](https://mlsanity.com/paper/1910.13027/) (2019)
- [Non-Stochastic Private Function Evaluation](https://mlsanity.com/paper/2010.09968/) (2020)

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

- [Differentially Private Synthetic Data via Foundation Model APIs 1: Images](https://mlsanity.com/paper/2305.15560/) (2023)
- [AdaPI: Facilitating DNN Model Adaptivity for Efficient Private Inference in Edge Computing](https://mlsanity.com/paper/2407.05633/) (2024)
- [Struct-Bench: A Benchmark for Differentially Private Structured Text Generation](https://mlsanity.com/paper/2509.10696/) (2025)
- [Privacy Measurement in Tabular Synthetic Data: State of the Art and Future Research Directions](https://mlsanity.com/paper/2311.17453/) (2023)
- [A supervised generative optimization approach for tabular data](https://mlsanity.com/paper/2309.05079/) (2023)
- [QuanText: Protecting Dataset-Level Secrets in Textual Data Sharing](https://mlsanity.com/paper/2609.17995/) (2026)

## Abstract (arXiv)

We study a setting where a data holder wishes to share data with a receiver, without revealing certain summary statistics of the data distribution (e.g., mean, standard deviation). It achieves this by passing the data through a randomization mechanism. We propose summary statistic privacy, a metric for quantifying the privacy risk of such a mechanism based on the worst-case probability of an adversary guessing the distributional secret within some threshold. Defining distortion as a worst-case Wasserstein-1 distance between the real and released data, we prove lower bounds on the tradeoff between privacy and distortion. We then propose a class of quantization mechanisms that can be adapted to different data distributions. We show that the quantization mechanism's privacy-distortion tradeoff matches our lower bounds under certain regimes, up to small constant factors. Finally, we demonstrate on real-world datasets that the proposed quantization mechanisms achieve better privacy-distortion tradeoffs than alternative privacy mechanisms.