BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Minimax rate for multivariate data under componentwise local diffe
 rential privacy constraints - Chiara Amorino (Universitat Pompeu Fabra)
DTSTART:20260220T140000Z
DTEND:20260220T150000Z
UID:TALK243493@talks.cam.ac.uk
CONTACT:Po-Ling Loh
DESCRIPTION:Our research analyses the balance between maintaining privacy 
 and preserving statistical accuracy when dealing with multivariate data th
 at is subject to componentwise local differential privacy (CLDP). With CLD
 P\, each component of the private data is made public through a separate p
 rivacy channel. This allows for varying levels of privacy protection for d
 ifferent components or for the privatization of each component by differen
 t entities\, each with their own distinct privacy policies. It also covers
  the practical situations where it is impossible to privatize jointly all 
 the components of the raw data. We develop general techniques for establis
 hing minimax bounds that shed light on the statistical cost of privacy in 
 this context\, as a function of the privacy levels $\\alpha_1\, \\dots \, 
 \\alpha_d$ of the $d$ components and demonstrate the versatility and effic
 iency of these techniques by presenting various statistical applications. 
 Additionally\, we conduct a detailed analysis of the effective privacy lev
 el\, exploring how information about a private characteristic of an indivi
 dual may be inferred from the publicly visible characteristics of the same
  individual.\n\nThe talk is based on a joint work with A. Gloter.\n\n*This
  talk is co-hosted by the Informed-AI Hub.*
LOCATION:MR12\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR