BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Distribution-Free\, Risk-Controlling Prediction Sets - Stephen Bat
 es\, University of California Berkeley
DTSTART:20210205T160000Z
DTEND:20210205T170000Z
UID:TALK156541@talks.cam.ac.uk
CONTACT:Dr Sergio Bacallado
DESCRIPTION:To enable valid statistical inference in prediction tasks\, we
  show how to generate set-valued predictions for black-box predictors that
  control the expected loss on future test points at a user-specified level
 . Our approach provides explicit finite-sample guarantees for any distribu
 tion by using a holdout set to calibrate the size of the prediction sets\,
  generalizing conformal prediction to control more complex notions of erro
 r such as the false rejection rate. We demonstrate our procedure in five l
 arge-scale problems: (1) classification problems where some mistakes are m
 ore costly than others\; (2) multi-label classification\, where each obser
 vation has multiple associated labels\; (3) classification problems where 
 the labels have a hierarchical structure\; (4) image segmentation\, where 
 we wish to predict a set of pixels containing an object of interest\; and 
 (5) protein structure prediction.
LOCATION: https://maths-cam-ac-uk.zoom.us/j/92821218455?pwd=aHFOZWw5bzVReU
 NYR2d5OWc1Tk15Zz09
END:VEVENT
END:VCALENDAR