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 (ε, δ)-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.
[ bib | video | poster | .pdf ] Back
This file was generated by bibtex2html 1.92.