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:Level 2 Lobby
DTSTART;TZID=America/Los_Angeles:20240626T180000
DTEND;TZID=America/Los_Angeles:20240626T190000
UID:dac_DAC 2024_sess256_LBR117@linklings.com
SUMMARY:Late Breaking Result: AQFP-aware Binary Neural Network Architectur
 e Search
DESCRIPTION:Late Breaking Results Poster\n\nZhengang Li (Northe), Xuan She
 n (Northeastern University), Geng Yuan (University of Georgia), Maoud Zabi
 hi (Northeastern University), Tomoharu Yamauchi (Tokyo City University), Y
 anzhi Wang (Northeastern University), and Olivia Chen (Tokyo City Universi
 ty)\n\nAdiabatic Quantum-Flux-Parametron (AQFP) is a superconducting logic
  with extremely high energy efficiency. Recent research has made initial s
 trides toward developing an AQFP-based crossbar accelerator. However sever
 al critical challenges from both the hardware and software side remain, pr
 eventing the design from being a comprehensive solution.\n\nIn this paper,
  we propose an AQFP-aware binary neural network architecture search framew
 ork that leverages software-hardware co-optimization to eventually search 
 the AQFP-adapted neural network and the corresponding hardware configurati
 on, providing a feasible AQFP-based solution for binary neural network (BN
 N) acceleration. Experimental results show that our framework can effectiv
 ely search the AQFP-adapted neural network, consistently outperforming the
  representative AQFP-based framework.
END:VEVENT
END:VCALENDAR