Fast Estimation for Privacy and Utility in Differentially Private Machine Learning

Recently, differential privacy has been widely studied in machine learning due to its formal privacy guarantees for data analysis. As one of the most important parameters of differential privacy, ϵ controls the crucial tradeoff between the strength of the privacy guarantee and the utility of model. Therefore, the choice of ϵ has a great influence on the performance of differentially private learning models. But so far, there is still no rigorous method for choosing ϵ. In this paper, we deduce the influence of ϵ on utility private learning models through strict mathematical derivation, and propose a novel approximate approach for estimating the utility of any ϵ value. We show that our approximate approach has a fairly small error and can be used to estimate the optimal ϵ according to the expected utility of users. Experimental results demonstrate high estimation accuracy and broad applicability of our approximate approach.