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:20211208T164408Z
LOCATION:3014
DTSTART;TZID=America/Los_Angeles:20211207T161000
DTEND;TZID=America/Los_Angeles:20211207T163000
UID:dac_DAC 2021_sess110_RESEARCH555@linklings.com
SUMMARY:COSAIM: Counter-based Stochastic-behaving Approximate Integer Mult
 iplier for Deep Neural Networks
DESCRIPTION:Research Manuscript\n\nCOSAIM: Counter-based Stochastic-behavi
 ng Approximate Integer Multiplier for Deep Neural Networks\n\nYu\n\nWe pro
 pose a new stochastic-behaving approximate integer unsigned multiplier, CO
 SAIM. The new design is an improved stochastic multiplier. Our evaluation 
 shows that the COSAIM with error improvement can achieve error bias, mean 
 error, peak errors at 0.06%, 0.3%, 1.81%. Experimental results show that C
 OSAIM can save up to 53.95%, 32.84%, 52.24%, 21.05% in area, power, energy
  and the product Area•1/Throughput, respectively. By doing parallel design
 , COSAIM can further lead to improvements in area, power and energy reduct
 ion by 60.44%, 53.33% and 68.54%, respectively. When implemented in CNN ap
 plication, COSAIM can achieve similar inference accuracy compared to the b
 aseline.\n\nTopic: Design\n\nKeyword: Approximate Computing for AI/ML\n\nV
 irtual Program: Presented In-Person
END:VEVENT
END:VCALENDAR