BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.4.0.1//EN
TZID:Europe/Paris
X-WR-TIMEZONE:Europe/Paris
BEGIN:VEVENT
UID:135@isdm.umontpellier.fr
DTSTART;TZID=Europe/Paris:20250306T153000
DTEND;TZID=Europe/Paris:20250306T153000
DTSTAMP:20260602T130232Z
URL:https://isdm.umontpellier.fr/events/how-should-we-construct-prediction
 -sets-insights-from-conformal-prediction-2/
SUMMARY:How Should We Construct Prediction Sets? Insights from Conformal Pr
 ediction
DESCRIPTION:Campus St Priest (860 Rue Saint Priest 34095 Montpellier Cedex 
 5)\, bat. 5\, Room: 02.124\nMachine Learning in Montpellier\, Theory &amp\
 ; Practice\nTiffany Ding (UC Berckley)\n\nIn the first part of the talk\, 
 I will present some reflections on the purpose of prediction sets and the 
 role that statistics can play in forming useful prediction sets. In partic
 ular\, I will discuss how prediction sets fit into a decision making pipel
 ine and the different kinds of decisions one may make using a prediction s
 et. In the second part of the talk\, I will describe a particular statisti
 cally motivated set-generating procedure for the classification setting ca
 lled clustered conformal prediction\, which gives all classes an equal cha
 nce of being correctly included in the prediction set (“class-conditiona
 l coverage”). This procedure can be useful in situations where it is imp
 ortant to identify instances of all classes\, even the rare ones. We demon
 strate the performance of this method on ImageNet and other image classifi
 cation datasets.\n\nLabéliséHallesIA
ATTACH;FMTTYPE=image/jpeg:https://isdm.umontpellier.fr/wp-content/uploads/
 2025/02/ml-mpt-1.jpg
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20241027T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR