BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160010Z
LOCATION:Track 10
DTSTART;TZID=America/New_York:20201119T100000
DTEND;TZID=America/New_York:20201119T103000
UID:submissions.supercomputing.org_SC20_sess388_cov101@linklings.com
SUMMARY:Enabling Rapid COVID-19 Small Molecule Drug Design Through Scalabl
 e Deep Learning of Generative Models
DESCRIPTION:ACM Gordon Bell COVID Finalist, Awards Presentation\n\nEnablin
 g Rapid COVID-19 Small Molecule Drug Design Through Scalable Deep Learning
  of Generative Models\n\nJacobs, Moon, McLoughlin, Jones, Hysom...\n\nWe i
 mproved the quality and reduced the time to produce machine-learned models
  for use in small molecule antiviral design. Our globally asynchronous mul
 ti-level parallel training approach strong scales to all of Sierra with up
  to 97.7% efficiency. We trained a novel, character-based Wasserstein auto
 encoder that produces a higher quality model trained on 1.613 billion comp
 ounds in 23 minutes while the previous state-of-the-art takes a day on 1 m
 illion compounds. Reducing training time from a day to minutes shifts the 
 model creation bottleneck from computer job turnaround time to human innov
 ation time. Our implementation achieves 318 PFLOPS for 17.1% of half-preci
 sion peak. We will incorporate this model into our molecular design loop, 
 enabling the generation of more diverse compounds: searching for novel, ca
 ndidate antiviral drugs improves and reduces the time to synthesize compou
 nds to be tested in the lab.\n\nRegistration Category: Tech Program Reg Pa
 ss
END:VEVENT
END:VCALENDAR