BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Supervised Learning with Missing Values
DTSTART:20230209T100000
DTEND:20230209T110000
DTSTAMP:20260407T120328Z
UID:4ad6c9ef8b4c7bb7a63010d35158778f38a68749876143d9d7790f88
CATEGORIES:Conferences - Seminars
DESCRIPTION:Julie JOSSE\, Inria \nSeminar in Mathematics\nAbstract: Missin
 g data is a pervasive issue in statistical practice\, appearing for a rang
 e of reasons such as device failure\, participant not answering sensitive 
 questions in polls\, and ever-increasing data volume. In this presentation
 \, I will briefly review the literature addressing missing data in an infe
 rential framework\, and then discuss the challenges posed by missing value
 s in supervised learning. This includes an analysis of the impute-then-reg
 ress procedure's consistency when using powerful non-parametric methods su
 ch as random forest\, as well as a description of the NeuMiss neural netwo
 rk architecture\, which allows for joint learning of imputation and regres
 sion. Finally\, I will illustrate the impact of the methods developed in t
 he causal inference field for estimating treatment effects from incomplete
  clinical data.\n 
LOCATION:MA A1 10 https://plan.epfl.ch/?room==MA%20A1%2010 https://epfl.zo
 om.us/j/62595185161
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR