BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CREST - ECPv5.1.3//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:CREST
X-ORIGINAL-URL:https://crest.science
X-WR-CALDESC:Events for CREST
BEGIN:VTIMEZONE
TZID:Europe/Paris
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:20190331T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:20191027T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Paris:20190527T140000
DTEND;TZID=Europe/Paris:20190527T151500
DTSTAMP:20260713T081546
CREATED:20190315T131105Z
LAST-MODIFIED:20190315T131105Z
UID:12232-1558965600-1558970100@crest.science
SUMMARY:Geoffrey CHINOT (CREST) - "ERM and RERM are tractable and optimal estimators in the ε-contamination model"
DESCRIPTION:\nThe Statistical Seminar: Every Monday at 2:00 pm.\nTime: 2:00 pm – 3:15 pm\nDate: 27 th of May 2019\nPlace: Room 3001.\nGeoffrey CHINOT (CREST) – “ERM and RERM are tractable and optimal estimators in the ε-contamination model” \nAbstract: We study the ERM and RERM with Lipschitz and convex loss functions under a subgaussian assumption on the design. We consider a setting where |O| outliers may contaminate the labels. In that case\, we show that the error rate is bounded by r(N) + |O|/N\, where N is the total number of observations and r(N) is the optimal error rate in the non-contaminated setting. The main result can be used for both non-regularized and regularized procedures. For instance we present results for the Huber’s M-estimators without penalization or regularized by the l1-norm. For these two applications we get minimax-optimal rates in the ε- contamination model. \n \nOrganizers:\nCristina BUTUCEA\, Alexandre TSYBAKOV\, Julie JOSSE\, Eric MOULINES\, Mathieu ROSENBAUM\nSponsors:\nCREST-CMAP\n \n\n
URL:https://crest.science/event/geoffrey-chinot-crest/
CATEGORIES:Statistics
ATTACH;FMTTYPE=:
END:VEVENT
END:VCALENDAR