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:360@isdm.umontpellier.fr
DTSTART;TZID=Europe/Paris:20250403T140000
DTEND;TZID=Europe/Paris:20250403T140000
DTSTAMP:20260828T094205Z
URL:https://isdm.umontpellier.fr/events/on-volume-minimization-in-conforma
 l-regression-2/
SUMMARY:On Volume Minimization in Conformal Regression
DESCRIPTION:Room 02.124\, Building 5\, St Priest campus\n\nMachine Learning
  in Montpellier\, Theory &amp\; Practice - Batiste Le Bars (Inria)\n\nConf
 ormal Prediction has recently been considered as one of the state-of-art t
 echnique to construct distribution-free prediction sets satisfying probabi
 listic coverage guarantees. In this talk\, we study the question of volume
  optimality in split conformal regression. Using the fact that the calibra
 tion step can be seen as an empirical volume minimization problem\, we fir
 st derive a finite-sample upper-bound on the excess volume loss of the int
 erval returned by the classical split method. This quantity measures the d
 ifference in length between the interval obtained with the split method an
 d the shortest oracle prediction interval. Then\, we introduce EffOrt\, a 
 methodology that modifies the learning step so that the base prediction fu
 nction minimizes the length of the returned intervals. In particular\, our
  theoretical analysis of the excess volume loss of the prediction sets pro
 duced by EffOrt reveals the links between the learning and calibration ste
 ps\, and notably the impact of the function class of the base predictor. W
 e also introduce Ad-EffOrt\, an extension of the previous method\, which p
 roduces intervals whose size adapts to the value of the covariate.\n\nMach
 ine Learning in Montpellier\, Theory &amp\; Practice
ATTACH;FMTTYPE=image/jpeg:https://isdm.umontpellier.fr/wp-content/uploads/
 2026/06/ml-mtp-gC78d5.png
CATEGORIES:ML MTP
LOCATION:Room 02.124 Building 5\, St Priest Campus\, Montpellier\, 34000\, 
 France
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=St Priest Campus\, Montpell
 ier\, 34000\, France;X-APPLE-RADIUS=100;X-TITLE=Room 02.124 Building 5:geo
 :0,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:DAYLIGHT
DTSTART:20250330T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR