BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Data perturbation for data science - Richard Samworth (University 
 of Cambridge)
DTSTART:20180629T100000Z
DTEND:20180629T104500Z
UID:TALK107524@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:When faced with a dataset and a problem of interest\, should w
 e propose a statistical model and use that to inform an appropriate algori
 thm\, or dream up a potential algorithm and then seek to justify it?  The 
 former is the more traditional statistical approach\, but the latter appea
 rs to be becoming more popular.  I will discuss a class of algorithms that
  belong in the second category\, namely those that involve data perturbati
 on (e.g. subsampling\, random projections\, artificial noise\, knockoffs\,
 ...).  As examples\, I will consider Complementary Pairs Stability Selecti
 on for variable selection and sparse PCA via random projections.  This wil
 l involve joint work with Rajen Shah\, Milana Gataric and Tengyao Wang.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR