FriendlyCore: Practical Differentially Private Aggregation

Differentially private algorithms for common metric aggregation tasks, such as clustering or averaging, often have limited practicality due to their complexity or a large number of data points that is required for accurate results. We propose a simple and practical tool FriendlyCore that takes a set of points D from an unrestricted (pseudo) metric space as input. When D has effective diameter r, FriendlyCore returns a “stable” subset DG ⊆ D that includes all points, except possibly few outliers, and is certified to have diameter r. FriendlyCore can be used to preprocess the input before privately aggregating it, potentially simplifying the aggregation or boosting its accuracy. Surprisingly, FriendlyCore is light-weight with no dependence on the dimension. We empirically demonstrate its advantages in boosting the accuracy of mean estimation, outperforming tailored methods.