BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Random projection ensemble classification - Timothy Cannings
DTSTART:20160128T143000Z
DTEND:20160128T160000Z
UID:TALK64177@talks.cam.ac.uk
CONTACT:Yingzhen Li
DESCRIPTION:We introduce a very general method for high-dimensional classi
 fication\, based on careful combination of the results of applying an arbi
 trary base classifier to random projections of the feature vectors into a 
 lower-dimensional space.  In one special case presented here\, the random 
 projections are divided into non-overlapping blocks\, and within each bloc
 k we select the projection yielding the smallest estimate of the test erro
 r.  Our random projection ensemble classifier then aggregates the results 
 of applying the base classifier on the selected projections\, with a data-
 driven voting threshold to determine the final assignment.  Our theoretica
 l results elucidate the effect on performance of increasing the number of 
 projections.  Moreover\, under a boundary condition implied by the suffici
 ent dimension reduction assumption\, we control the test excess risk of th
 e random projection ensemble classifier.  A simulation comparison with sev
 eral other popular high-dimensional classifiers reveals its excellent fini
 te-sample performance.  This is joint work with Richard Samworth.\n\nPaper
 : http://arxiv.org/abs/1504.04595\n
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR