Composition Properties of Inferential Privacy for Time-Series Data

TL;DR

This paper investigates how a specific form of inferential privacy, called Pufferfish, composes over time-series data, and finds that the Markov Quilt Mechanism offers strong composition guarantees similar to differential privacy.

Contribution

It demonstrates that the Markov Quilt Mechanism for Pufferfish privacy has strong composition properties for time-series data, addressing a key barrier in practical inferential privacy applications.

Findings

Markov Quilt Mechanism has strong composition properties.

Pufferfish mechanisms may not compose gracefully in general.

The study bridges a gap between inferential privacy and differential privacy for time series.

Abstract

With the proliferation of mobile devices and the internet of things, developing principled solutions for privacy in time series applications has become increasingly important. While differential privacy is the gold standard for database privacy, many time series applications require a different kind of guarantee, and a number of recent works have used some form of inferential privacy to address these situations. However, a major barrier to using inferential privacy in practice is its lack of graceful composition -- even if the same or related sensitive data is used in multiple releases that are safe individually, the combined release may have poor privacy properties. In this paper, we study composition properties of a form of inferential privacy called Pufferfish when applied to time-series data. We show that while general Pufferfish mechanisms may not compose gracefully, a specific Pufferfish mechanism, called the Markov Quilt Mechanism, which was recently introduced, has strong composition properties comparable to that of pure differential privacy when applied to time series data.