BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Selecting Groups of Variables for Prediction Problems in Chemometr
 ics:  Recent Regularization Approaches - Rainer von Sachs (Université Cat
 holique de Louvain )
DTSTART:20180116T144500Z
DTEND:20180116T153000Z
UID:TALK97642@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:This presentation addresses the problem of selecting important
 \, potentially overlapping groups of predictor variables  in linear models
  such that the resulting model satisfies a balance between interpretabilit
 y and prediction performance.  This is motivated by data from the field of
  chemometrics where\, due to correlation between predictors from different
   groups (i.e. variable group &ldquo\;overlap&rdquo\;)\, identifying group
 s during model estimation is particularly challenging.  In particular\, we
  will highlight some issues of existing methods when they are applied to h
 igh dimensional data with  overlapping groups of variables. This will be d
 emonstrated through comparison of their optimization criteria and  their p
 erformance on simulated data.   This is joint work in progress with Rebecc
 a Marion\, ISBA\, Universit&eacute\; catholique de Louvain\, Belgium.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR