BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards Meaningful Stochastic Defences in Machine Learning - Ilia 
 Shumailov\, University of Oxford
DTSTART:20221115T140000Z
DTEND:20221115T150000Z
UID:TALK184949@talks.cam.ac.uk
CONTACT:Kieron Ivy Turk
DESCRIPTION:Machine learning (ML) has proven to be more fragile than previ
 ously thought\, especially in adversarial settings. A capable adversary ca
 n cause ML systems to break at training\, inference\, and deployment stage
 s. In this talk\, I will cover the recent work on attacking and defending 
 machine learning pipelines using stochastic defences\; I will describe how
 \, seemingly powerful defences fail to provide any security and end up bei
 ng vulnerable to even standard attackers. I will then demonstrate a number
  of possible randomness-based defences that can provide theoretical and pr
 actical performance improvements.\n\nBio: Ilia Shumailov holds a PhD in Co
 mputer Science from University of Cambridge\, specialising in Machine Lear
 ning and Computer Security. During the PhD under the supervision of Prof R
 oss Anderson Ilia has worked on a number of projects spanning the fields o
 f machine learning security\, cybercrime analysis and signal processing. F
 ollowing the PhD\, Ilia joined Vector Institute in Canada as a Postdoctora
 l Fellow\, where he worked under the supervision of Prof Nicolas Papernot 
 and Prof Kassem Fawaz. Ilia is currently a Junior Research Fellow at Chris
 t Church\, University of Oxford. 
LOCATION:Webinar &amp\; FW11\, Computer Laboratory\, William Gates Buildin
 g.
END:VEVENT
END:VCALENDAR