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/Helsinki
BEGIN:DAYLIGHT
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:EEST
DTSTART:20230326T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
TZNAME:EET
DTSTART:20231029T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/Helsinki:20230413T141500
DTEND;TZID=Europe/Helsinki:20230413T151500
DTSTAMP:20260711T074237
CREATED:20230407T081531Z
LAST-MODIFIED:20230407T081531Z
UID:14857-1681395300-1681398900@crest.science
SUMMARY:Matus TELGARSKY (Université Illinois\, Urbana Champaign) - "Searching for the implicit bias of deep learning"
DESCRIPTION:Statistical Seminar: \nTime: 2:15 pm – 3:15 pm\nDate: 13th of April 2023\nPlace: Room 3001 + ZOOM \n  \nMatus TELGARSKY (Université Illinois\, Urbana Champaign) – “Searching for the implicit bias of deep learning” \n  \nAbstract: \nWhat makes deep learning special — why is it effective in so many settings where other models fail? This talk will present recent progress from three perspectives. The first result is approximation-theoretic: deep networks can easily represent phenomena that require exponentially-sized shallow networks\, decision trees\, and other classical models. Secondly\, I will show that their statistical generalization ability — namely\, their ability to perform well on unseen testing data — is correlated with their prediction margins\, a classical notion of confidence. Finally\, comprising the majority of the talk\, I will discuss the interaction of the preceding two perspectives with optimization: specifically\, how standard descent methods are implicitly biased towards models with good generalization. Here I will present two approaches: the strong implicit bias\, which studies convergence to specific well-structured objects\, and the weak implicit bias\, which merely ensures certain good properties eventually hold\, but has a more flexible proof technique. \n  \n  \nLink :  https://zoom.us/j/91051481144?pwd=alF1cjJUZ0pmUlprRmJjUWRDNU9odz09   \nID de réunion : 910 5148 1144\nCode secret : 590133 \n  \nOrganizers:\nCristina BUTUCEA (CREST)\, Alexandre TSYBAKOV (CREST)\, Karim LOUNICI (CMAP) \, Jaouad MOURTADA (CREST)\nSponsors:\nCREST-CMAP \n
URL:https://crest.science/event/matus-telgarsky-universite-illinois-urbana-champaign-searching-for-the-implicit-bias-of-deep-learning/
CATEGORIES:Statistics
ATTACH;FMTTYPE=:
END:VEVENT
END:VCALENDAR