Deciding Accuracy of Differential Privacy Schemes
This work addresses a foundational gap in differential privacy theory, offering a unified accuracy framework that could impact privacy-preserving algorithm design, though it is incremental in refining theoretical foundations.
The paper tackles the lack of a general definition for accuracy in differential privacy by introducing a new notion based on program discontinuity and distance, which subsumes existing definitions and sometimes yields better claims. It shows accuracy is generally undecidable but identifies a class of computations where it is decidable, implementing a procedure to generate proofs or counterexamples for common algorithms.
Differential privacy is a mathematical framework for developing statistical computations with provable guarantees of privacy and accuracy. In contrast to the privacy component of differential privacy, which has a clear mathematical and intuitive meaning, the accuracy component of differential privacy does not have a generally accepted definition; accuracy claims of differential privacy algorithms vary from algorithm to algorithm and are not instantiations of a general definition. We identify program discontinuity as a common theme in existing \emph{ad hoc} definitions and introduce an alternative notion of accuracy parametrized by, what we call, {\distance} -- the {\distance} of an input $x$ w.r.t., a deterministic computation $f$ and a distance $d$, is the minimal distance $d(x,y)$ over all $y$ such that $f(y)\neq f(x)$. We show that our notion of accuracy subsumes the definition used in theoretical computer science, and captures known accuracy claims for differential privacy algorithms. In fact, our general notion of accuracy helps us prove better claims in some cases. Next, we study the decidability of accuracy. We first show that accuracy is in general undecidable. Then, we define a non-trivial class of probabilistic computations for which accuracy is decidable (unconditionally, or assuming Schanuel's conjecture). We implement our decision procedure and experimentally evaluate the effectiveness of our approach for generating proofs or counterexamples of accuracy for common algorithms from the literature.