BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Composition in Differential Privacy - Dr Cynthia Dwork\, Microsoft
  Research
DTSTART:20161201T174500Z
DTEND:20161201T190000Z
UID:TALK69433@talks.cam.ac.uk
CONTACT:Kasia Warburton
DESCRIPTION:A signal strength of Differential Privacy -- a mathematically\
 nrigorous definition of privacy tailored to large datasets -- is the\nabil
 ity not just to quantify privacy loss but also to analyze and\ncontrol cum
 ulative loss over multiple computations. In cryptographic\nterms\, we unde
 rstand how privacy loss "composes". Not only is this\ncrucial for the real
 -word use of data but it also gives rise to a rich\nalgorithmic literature
 \, as simple differentially private primitives can\nbe combined in creativ
 e ways to construct complex differentially private\nanalyses. This talk di
 scusses recent analytical results in the\ncomposition of differentially pr
 ivate algorithms.\n
LOCATION:Mill Lane Lecture Room 1\, 8 Mill Lane\, Cambridge CB2 1RX
END:VEVENT
END:VCALENDAR