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:20240626T180033Z
LOCATION:3008\, 3rd Floor
DTSTART;TZID=America/Los_Angeles:20240625T160000
DTEND;TZID=America/Los_Angeles:20240625T161500
UID:dac_DAC 2024_sess109_RESEARCH1464@linklings.com
SUMMARY:MTL-Split: Multi-Task Learning for Edge Devices using Split Comput
 ing
DESCRIPTION:Research Manuscript\n\nLuigi Capogrosso and Enrico Fraccaroli 
 (University of Verona); Samarjit Chakraborty (University of North Carolina
 , Chapel Hill); and Franco Fummi and Marco Cristani (University of Verona)
 \n\nSplit Computing (SC), where a Deep Neural Network (DNN) is intelligent
 ly split with a part of it deployed on an edge device and the rest on a re
 mote server is emerging as a promising approach. It allows the power of DN
 Ns to be leveraged for latency-sensitive applications that do not allow th
 e entire DNN to be deployed remotely, while not having sufficient computat
 ion bandwidth available locally. In many such embedded scenarios, such as 
 those in the automotive domain, computational resource constraints also ne
 cessitate MultiTask Learning (MTL), where the same DNN is used for multipl
 e inference tasks instead of having dedicated DNNs for each task, which wo
 uld need more computing bandwidth. However, how to partition such a multi-
 tasking DNN to be deployed within a SC framework has not been sufficiently
  studied. This paper studies this problem and MTL-Split, our novel propose
 d architecture, shows encouraging results on both synthetic and real-world
  data. The code implementing this architecture will be made publicly avail
 able.\n\nTopic: Design\n\nKeyword: Design of Cyber-physical Systems and Io
 T\n\nSession Chairs: Chen Liu (Intel Corporation) and Jakub Szefer (Yale U
 niversity)
END:VEVENT
END:VCALENDAR