BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards multi-purpose locally differentially-private synthetic dat
 a release via plug-in estimation - Botond Szabo (Bocconi University)
DTSTART:20260213T140000Z
DTEND:20260213T150000Z
UID:TALK243616@talks.cam.ac.uk
CONTACT:Po-Ling Loh
DESCRIPTION:We develop plug-in estimators for locally differentially priva
 te semi-parametric estimation via spline wavelets. The approach leads to o
 ptimal rates of convergence for a large class of estimation problems that 
 are characterized by (differentiable) functionals $\\Lambda(f)$ of the tru
 e data generating density $f$. The crucial feature of the locally private 
 data $Z_1\,\\dots\, Z_n$ we generate is that it does not depend on the par
 ticular functional $\\Lambda$ (or the unknown density $f$) the analyst wan
 ts to estimate. Hence\, the synthetic data can be generated and stored a p
 riori and can subsequently be used by any number of analysts to estimate m
 any vastly different functionals of interest at the provably optimal rate.
  In principle\, this removes a long standing practical limitation in stati
 stics of differential privacy\, namely\, that optimal privacy mechanisms n
 eed to be tailored towards the specific estimation problem at hand.\n\n*Th
 is talk is co-hosted by the Informed-AI Hub.*
LOCATION:MR12\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR