Sample-and-threshold differential privacy: Histograms and applications

Federated analytics aims to compute accurate statistics from distributed datasets. A "Differential Privacy" (DP) guarantee is usually desired by the users of the devices storing the data. In this work, we prove a strong $(\epsilon, \delta)$-DP guarantee for a highly practical sampling-based procedure to derive histograms. We also provide accuracy guarantees and show how to apply the procedure to estimate quantiles and modes.