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:20211208T164413Z
LOCATION:3014
DTSTART;TZID=America/Los_Angeles:20211209T133000
DTEND;TZID=America/Los_Angeles:20211209T135300
UID:dac_DAC 2021_sess118_RESEARCH653@linklings.com
SUMMARY:Towards Resilient Deployment of In-Memory Neural Networks with Hig
 h Throughput
DESCRIPTION:Research Manuscript\n\nTowards Resilient Deployment of In-Memo
 ry Neural Networks with High Throughput\n\nZhang\n\nResistive computing sy
 stems (RCSs) promise exascale computing capabilities to inference engines 
 for deep learning. In this paper, we propose a framework for resilient dep
 loyment of high throughput CNNs to RCSs.  The framework is based on integr
 ating a data layout organization step and a distribution guided training s
 tep into the flow for mapping CNNs to RCSs.  The experimental results demo
 nstrate that the proposed techniques expand the average solution space for
  data layout organization with 1.4* 10^14X. This translates into that CNNs
  with high throughput can be deployed onto RCS with up to 10% defects and 
 still attain high classification accuracy.\n\nTopic: Design\n\nKeyword: AI
 /ML System Design\n\nVirtual Program: Presented In-Person
END:VEVENT
END:VCALENDAR