Optimal Differentially Private Mechanisms for Randomised Response
This work provides incremental improvements in privacy-preserving data analysis by optimizing mechanisms for a specific technique.
The authors tackled the problem of finding optimal differentially private mechanisms for randomised response, presenting closed-form solutions for both strict and relaxed privacy definitions in binary output settings, including the original Warner method.
We examine a generalised Randomised Response (RR) technique in the context of differential privacy and examine the optimality of such mechanisms. Strict and relaxed differential privacy are considered for binary outputs. By examining the error of a statistical estimator, we present closed solutions for the optimal mechanism(s) in both cases. The optimal mechanism is also given for the specific case of the original RR technique as introduced by Warner in 1965.