BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Introduction to differential privacy - Talay Cheema &amp\; Ferenc 
 Huszar (University of Cambridge)
DTSTART:20220309T110000Z
DTEND:20220309T123000Z
UID:TALK171488@talks.cam.ac.uk
CONTACT:Elre Oldewage
DESCRIPTION:Privacy is increasingly important to society and coming under 
 more scrutiny. Modern machine learning methods consume vast amounts of dat
 a\, and it is well known that many large networks memorise their training 
 data. It is necessary to move beyond ineffective heuristics such as anonym
 isation or selective withholding of data. Differential privacy provides a 
 framework for quantifying loss of privacy and designing algorithms which k
 eep reasonable worst case privacy loss within acceptable levels. In this t
 alk\, we will motivate and introduce foundational differential privacy met
 hods\, and look at applications to machine learning.\n\n \n\nRecommended r
 eading: Chapter 1 of The Algorithmic Foundations of Differential Privacy\;
  Cynthia Dwork and Aaron Roth (2014) ("privacybook.pdf (upenn.edu)":https:
 //www.cis.upenn.edu/~aaroth/Papers/privacybook.pdf)\n\n\nSuggested reading
 : Deep Learning with Differential Privacy\; Abadi et al (2016) ("1607.0013
 3.pdf (arxiv.org)":https://arxiv.org/pdf/1607.00133.pdf)\n\nOur reading gr
 oups are live-streamed via Zoom and recorded for our Youtube channel. The 
 Zoom details are distributed via our weekly mailing list. *Please note the
  change in venu.*
LOCATION: Cambridge University Engineering Department\, CBL Seminar room B
 E4-38
END:VEVENT
END:VCALENDAR