BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260522T150115Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181113T083000
DTEND;TZID=America/Chicago:20181113T170000
UID:submissions.supercomputing.org_SC18_sess322_post153@linklings.com
SUMMARY:AI Matrix – Synthetic Benchmarks for DNN
DESCRIPTION:Wei Wei, Lingjie Xu, Lingling Jin, and Wei Zhang (Alibaba Inc)
  and Tianjun Zhang (University of California, Berkeley)\n\nThe current AI 
 benchmarks suffer from a number of drawbacks. First, they cannot adapt to 
 the emerging changes of deep learning (DL) algorithms and are fixed once s
 elected. Second, they contain tens to hundreds of applications and have ve
 ry long running time. Third, they are mainly selected from open sources, w
 hich are restricted by copyright and not representable of the proprietary 
 applications. To address these drawbacks, this work firstly proposes a syn
 thetic benchmark framework that generates a small number of benchmarks tha
 t best represent a broad range of applications using their profiled worklo
 ad characteristics. The synthetic benchmarks can adapt to new DL algorithm
 s by re-profiling new applications and updating itself, greatly reduce num
 ber of benchmark tests and running time, and strongly represent DL applica
 tions of interests. The framework is validated by using log data profiled 
 from DL models running on Alibaba AI platform, and is representable of rea
 l workload characteristics.\n\nRegistration Category: Tech Program Reg Pas
 s, Exhibits Reg Pass\n\n
END:VEVENT
END:VCALENDAR