BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Vienna
BEGIN:DAYLIGHT
DTSTART:20210328T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20211031T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260403T220456Z
UID:1634133600@ist.ac.at
DTSTART:20211013T160000
DTEND:20211013T180000
DESCRIPTION:Speaker: Francis Bach\nhosted by Marco Mondelli\nAbstract: Many
  classification tasks in machine learning lie beyond the classical binary 
 and multi-class classification settings. In those tasks\, the output eleme
 nts are structured objects made of interdependent parts\, such as sequence
 s in natural language processing\, images in computer vision\, permutation
 s in ranking or matching problems\, etc. The structured prediction setting
  has two key properties that makes it radically different from multi-class
  classification\, namely\, the exponential growth of the size of the outpu
 t space with the number of its parts\, and the cost-sensitive nature of th
 e learning task\, as prediction mistakes are not equally costly. In this 
 talk\, I will present recent work on the design on loss functions that com
 bine numerical efficiency and statistical consistency (joint work with Ale
 ssandro Rudi\, Alex Nowak-Vila\, Vivien Cabannes).
LOCATION:Zoom Link: https://istaustria.zoom.us/j/95068695661?pwd=M2YwZUt1M1
 AvcXJlVUJjV1dBWDVEUT09\; Meeting ID: 950 6869 5661 Passcode: 741514\, ISTA
ORGANIZER:
SUMMARY:Francis Bach: Structured prediction: beyond support vector machine 
 and cross entropy
URL:https://talks-calendar.ista.ac.at/events/3270
END:VEVENT
END:VCALENDAR