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:20240626T180035Z
LOCATION:Level 2 Lobby
DTSTART;TZID=America/Los_Angeles:20240625T180000
DTEND;TZID=America/Los_Angeles:20240625T190000
UID:dac_DAC 2024_sess236_RESEARCH2027@linklings.com
SUMMARY:nvmXR: Design Space Exploration of Non-Volatile Memory Architectur
 es for Edge-XR Systems
DESCRIPTION:Work-in-Progress Poster\n\nZihan Zhang and Marco Donato (Tufts
  University)\n\nThe concurrent execution of deep neural network (DNN) mode
 ls over various input modalities is a growing trend for tackling complex s
 cenarios where user inputs and contextual features collected by sampling v
 isual and auditory signals are combined to provide more immersive and inte
 lligent user-computer interactions. One field in which these types of appl
 ication scenarios have shown to be pervasive is eXtended Reality (XR). One
  key challenge in enabling multi-modal execution is the requirement for XR
  systems to host and execute inference operations on a collection of diffe
 rent DNN models on mobile devices such as headsets. This work explores the
  optimization of XR memory architectures using embedded non-volatile memor
 ies (NVMs). We present nvmXR, a cross-stack evaluation framework that allo
 ws designers to compare different ReRAM-based architectures and identify o
 ptimal memory solutions based on specific workload constraints.\n\nTopic: 
 AI, Autonomous Systems, Cloud, Design, EDA, Embedded Systems, IP, Security
END:VEVENT
END:VCALENDAR