BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211208T164412Z
LOCATION:Level 2 - Lobby
DTSTART;TZID=America/Los_Angeles:20211208T180000
DTEND;TZID=America/Los_Angeles:20211208T190000
UID:dac_DAC 2021_sess243_WIP189@linklings.com
SUMMARY:CANCELLED: ARGOS: an Adaptive and ReGion-scale knowledge distillat
 ion for Object recognition Systems
DESCRIPTION:Networking Reception, Work-in-Progress Poster\n\nCANCELLED: AR
 GOS: an Adaptive and ReGion-scale knowledge distillation for Object recogn
 ition Systems\n\nFarhadi\n\nObject recognition research communities have a
 chieved great success using neural networks. Reducing these networks' comp
 utational cost remains a challenge, which impedes their deployment on embe
 dded devices. Recent studies employ a keyframe selection and temporal know
 ledge distillation for detecting objects in related frames to reduce these
  networks' computational cost. Still, keyframe selection models' processin
 g time substantially increases the system's response time. Consequently, w
 e present an online knowledge distillation framework (ARGOS) that improves
  the performance and the response time by decomposing input frame into a f
 ew sub-processes executed in parallel. Extensive experiments demonstrate i
 ts effectiveness with an average 89% reduction in energy.\n\nVirtual Progr
 am: Presented In-Person
END:VEVENT
END:VCALENDAR