BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//https://caida.ubc.ca//NONSGML iCalcreator 2.41.92//
CALSCALE:GREGORIAN
METHOD:PUBLISH
UID:30643938-3334-4866-a530-386262333662
X-WR-RELCALID:efc09d74-9c93-479e-a94f-485231ddccde
X-WR-TIMEZONE:America/Vancouver
X-WR-CALNAME:Contrastive Learning for ML-guided MIP search - Bistra Dilkina
 \, Associate Professor\, University of Southern California
BEGIN:VTIMEZONE
TZID:America/Vancouver
TZUNTIL:20261101T090000Z
BEGIN:STANDARD
TZNAME:PST
DTSTART:20241103T020000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
RDATE:20251102T020000
END:STANDARD
BEGIN:DAYLIGHT
TZNAME:PDT
DTSTART:20240310T020000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
RDATE:20250309T020000
RDATE:20260308T020000
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:c7816f99-7b5a-4799-817b-1538696698ad
DTSTAMP:20260726T220029Z
CLASS:PUBLIC
CREATED:20241217T162820Z
DESCRIPTION:Abstract: Recent research has demonstrated the ability to signi
 ficantly improve MIP solving by integrating ML-guided components and learn
 ing policies tailored to specific problem distributions. This benefits dir
 ectly the real-world deployment of Mixed Integer Programming models as the
 y are often used to solve repeatedly similar problems\, arising within a s
 pecific application context. We show that one important aspect of designin
 g an effective ML integration in MIP solving is the training loss utilized
  to tune the ML model parameters. In particular\, many important (heuristi
 c) tasks in MIP…
DTSTART;TZID=America/Vancouver:20241216T143000
DTEND;TZID=America/Vancouver:20241216T153000
LAST-MODIFIED:20241217T163334Z
LOCATION:UBC Vancouver Campus\, ICCS X836
SUMMARY:Contrastive Learning for ML-guided MIP search - Bistra Dilkina\, As
 sociate Professor\, University of Southern California
TRANSP:OPAQUE
URL:https://caida.ubc.ca/event/contrastive-learning-ml-guided-mip-search-bi
 stra-dilkina-associate-professor-university
END:VEVENT
END:VCALENDAR