BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Vienna
BEGIN:DAYLIGHT
DTSTART:20220327T030000
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:20260406T042317Z
UID:1644508800@ist.ac.at
DTSTART:20220210T170000
DTEND:20220210T180000
DESCRIPTION:Speaker: Virginia Smith\nhosted by Marco Mondelli\nAbstract: Ex
 ponential tilting is a technique commonly used to create parametric distri
 bution shifts. Despite its prevalence in related fields\, tilting has not 
 seen widespread use in machine learning. In this talk\, I discuss a simple
  extension to ERM---tilted empirical risk minimization (TERM)---which uses
  tilting to flexibly tune the impact of individual losses. I make connecti
 ons between TERM and related approaches\, such as Value-at-Risk\, Conditio
 nal Value-at-Risk\, and distributionally robust optimization (DRO)\, and p
 resent batch and stochastic first-order optimization methods for solving T
 ERM at scale. Finally\, I show that this baseline can be used for a multit
 ude of applications in machine learning\, such as enforcing fairness betwe
 en subgroups\, mitigating the effect of outliers\, and handling class imba
 lance---delivering state-of-the-art performance relative to more complex\,
  bespoke solutions for these problems. 
LOCATION:Zoom Link: https://istaustria.zoom.us/j/64653423567?pwd=eEYvUkJWY2
 VwdlBWMEFZT1BYTGdhUT09  Meeting ID: 646 5342 3567 Passcode: 469530\, ISTA
ORGANIZER:
SUMMARY:Virginia Smith: Tilted Losses in Machine Learning: Theory and Appli
 cations
URL:https://talks-calendar.ista.ac.at/events/3510
END:VEVENT
END:VCALENDAR