A Tight Context-aware Privacy Bound for Histogram Publication

2025-08-26

AI Summary

This paper addresses privacy guarantees for histogram publication under the Laplace mechanism. It proposes a context-aware privacy analysis framework based on pointwise maximal leakage (PML), departing from the distribution-agnostic assumptions of standard differential privacy. By incorporating prior knowledge of the true data distribution, the framework derives tighter privacy bounds: when histogram bin probabilities are bounded away from zero, strictly stronger privacy is achieved at the same noise level. Theoretical analysis and empirical evaluation demonstrate that the method significantly improves privacy precision without sacrificing utility, yielding a superior privacy–utility trade-off. The core contribution lies in systematically integrating data distribution knowledge into PML-based analysis, establishing a novel paradigm for context-sensitive privacy quantification.

Technology Category

• Machine Learning: Privacy
• Computer Vision: Bias, Fairness & Privacy
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments

Abstract

We analyze the privacy guarantees of the Laplace mechanism releasing the histogram of a dataset through the lens of pointwise maximal leakage (PML). While differential privacy is commonly used to quantify the privacy loss, it is a context-free definition that does not depend on the data distribution. In contrast, PML enables a more refined analysis by incorporating assumptions about the data distribution. We show that when the probability of each histogram bin is bounded away from zero, stronger privacy protection can be achieved for a fixed level of noise. Our results demonstrate the advantage of context-aware privacy measures and show that incorporating assumptions about the data can improve privacy-utility tradeoffs.

Problem

• Analyzing privacy guarantees of Laplace mechanism for histogram publication
• Incorporating data distribution assumptions via pointwise maximal leakage
• Improving privacy-utility tradeoffs through context-aware privacy measures

Innovation

• Uses pointwise maximal leakage for privacy analysis
• Incorporates data distribution assumptions for refined analysis
• Achieves stronger protection with bounded histogram probabilities