BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Europe/Skopje
BEGIN:DAYLIGHT
DTSTART:20240331T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241027T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240514T222229Z
UID:28D9704A-C2E1-4F9C-8DD7-EBD351AC0B45
DTSTART;TZID=Europe/Skopje:20240423T140000
DTEND;TZID=Europe/Skopje:20240423T153000
DESCRIPTION:Our reliance on smartphones demands continual advancement of mo
 bile computing. Yet\, our computing appetites grow much faster than the cu
 rrent hardware technology advances\, producing a critical strain on the mo
 bile’s limited resources. To address this issue\, we propose Approximate
  Mobile Computing (AMC) and take a radical stance that computation does no
 t need to be 100% precise. We first examine situations\, such as mobile vi
 deo playback and mobile deep learning for human activity recognition\, whe
 re the properties of the input and the limitations of human perception ope
 n space for AMC. We then develop methods that bring AMC to consumer device
 s\, including an Android compiler framework that enables dynamic tuning of
  the level of approximation according to the context of usage. Finally\, w
 e look into the future of AMC for efficient mobile sensing and model train
 ing as well.\n\nSpeaker(s): Veljko Pejovic\, PhD\, \n\nRoom: Meeting Room\
 , Bldg: FEEIT\, Rugjer Boshkovikj 18\, Postal box 574\, Skopje\, Macedonia
 \, Macedonia\, 1000
LOCATION:Room: Meeting Room\, Bldg: FEEIT\, Rugjer Boshkovikj 18\, Postal b
 ox 574\, Skopje\, Macedonia\, Macedonia\, 1000
ORGANIZER:katarina.trojacanec@finki.ukim.mk
SEQUENCE:65
SUMMARY:Invited Lecture: Bringing Fast Deep Learning to Mobiles through App
 roximate Computing
URL;VALUE=URI:https://events.vtools.ieee.org/m/417770
X-ALT-DESC:Description: <br /><p>Our reliance on smartphones demands contin
 ual advancement of&nbsp\;mobile computing. Yet\, our computing appetites g
 row much faster than the&nbsp\;current hardware technology advances\, prod
 ucing a critical strain on the&nbsp\;mobile&rsquo\;s limited resources. To
  address this issue\, we propose&nbsp\;Approximate Mobile Computing (AMC) 
 and take a radical stance that&nbsp\;computation does not need to be 100% 
 precise. We first examine&nbsp\;situations\, such as mobile video playback
  and mobile deep learning for&nbsp\;human activity recognition\, where the
  properties of the input and the&nbsp\;limitations of human perception ope
 n space for AMC. We then develop&nbsp\;methods that bring AMC to consumer 
 devices\, including an Android&nbsp\;compiler framework that enables dynam
 ic tuning of the level of&nbsp\;approximation according to the context of 
 usage. Finally\, we look into&nbsp\;the future of AMC for efficient mobile
  sensing and model training as well.</p>
END:VEVENT
END:VCALENDAR