BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//https://caida.ubc.ca//NONSGML iCalcreator 2.41.92//
CALSCALE:GREGORIAN
METHOD:PUBLISH
UID:34363063-6530-4530-b762-643534313265
X-WR-RELCALID:efc09d74-9c93-479e-a94f-485231ddccde
X-WR-TIMEZONE:America/Vancouver
X-WR-CALNAME:Robust Deep Learning Under Distribution Shift - Zachary Chase 
 Lipton\, Assistant Professor\, Carnegie Mellon University
BEGIN:VTIMEZONE
TZID:America/Vancouver
TZUNTIL:20211107T090000Z
BEGIN:STANDARD
TZNAME:PST
DTSTART:20191103T020000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
RDATE:20201101T020000
END:STANDARD
BEGIN:DAYLIGHT
TZNAME:PDT
DTSTART:20190310T020000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
RDATE:20200308T020000
RDATE:20210314T020000
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:03e8a7ed-1302-41d4-aa7e-ea3892b98d84
DTSTAMP:20261023T084952Z
CLASS:PUBLIC
CREATED:20191203T183127Z
DESCRIPTION:Abstract: We might hope that when faced with unexpected inputs\
 , well-designed software systems would fire off warnings. However\, ML sys
 tems\, which depend strongly on properties of their inputs (e.g. the i.i.d
 . assumption)\, tend to fail silently. Faced with distribution shift\, we 
 wish (i) to detect and (ii) to quantify the shift\, and (iii) to correct o
 ur classifiers on the fly—when possible. This talk will describe a line of
  recent work on tackling distribution shift. First\, I will focus on recen
 t work on label shift\, a classic problem\, where strong assumptions enabl
 e principled methods. Then I…
DTSTART;TZID=America/Vancouver:20191216T113000
DTEND;TZID=America/Vancouver:20191216T123000
LAST-MODIFIED:20210611T170858Z
LOCATION:ICCS - X836\, ICICS Computer Science\, 2366 Main Mall\, Vancouver\
 , BC
SUMMARY:Robust Deep Learning Under Distribution Shift - Zachary Chase Lipto
 n\, Assistant Professor\, Carnegie Mellon University
TRANSP:OPAQUE
URL:https://caida.ubc.ca/event/robust-deep-learning-under-distribution-shif
 t-zachary-chase-lipton-assistant-professor
END:VEVENT
END:VCALENDAR