Composition Theorems for f-Differential Privacy
By: Natasha Fernandes, Annabelle McIver, Parastoo Sadeghi
Potential Business Impact:
Protects your private information better online.
"f differential privacy" (fDP) is a recent definition for privacy privacy which can offer improved predictions of "privacy loss". It has been used to analyse specific privacy mechanisms, such as the popular Gaussian mechanism. In this paper we show how fDP's foundation in statistical hypothesis testing implies equivalence to the channel model of Quantitative Information Flow. We demonstrate this equivalence by a Galois connection between two partially ordered sets. This equivalence enables novel general composition theorems for fDP, supporting improved analysis for complex privacy designs.
Similar Papers
General-Purpose $f$-DP Estimation and Auditing in a Black-Box Setting
Cryptography and Security
Checks if private data is truly hidden.
Infinitely divisible privacy and beyond I: resolution of the $s^2=2k$ conjecture
Statistics Theory
Makes private data sharing safer with new math.
Differential privacy from axioms
Data Structures and Algorithms
Guarantees privacy even when attackers know almost everything.