BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20181221T160731Z
LOCATION:A2 Ballroom
DTSTART;TZID=America/Chicago:20181114T160000
DTEND;TZID=America/Chicago:20181114T163000
UID:submissions.supercomputing.org_SC18_sess466_gb103@linklings.com
SUMMARY:167-PFlops Deep Learning for Electron Microscopy: From Learning Ph
 ysics to Atomic Manipulation
DESCRIPTION:ACM Gordon Bell Finalist, Awards Presentation\n\n\n167-PFlops 
 Deep Learning for Electron Microscopy: From Learning Physics to Atomic Man
 ipulation\n\nPatton, Johnston, Young, Schuman, March...\n\nAn artificial i
 ntelligence system called MENNDL, which used 25,200 Nvidia Volta GPUs on O
 ak Ridge National Laboratory’s Summit machine, automatically designed an o
 ptimal deep learning network in order to extract structural information fr
 om raw atomic-resolution microscopy data. In a few hours, MENNDL creates a
 nd evaluates millions of networks using a scalable, parallel, asynchronous
  genetic algorithm augmented with a support vector machine to automaticall
 y find a superior deep learning network topology and hyper-parameter set t
 han a human expert can find in months. For the application of electron mic
 roscopy, the system furthers the goal of improving our understanding of th
 e electron-beam-matter interactions and real-time image-based feedback, wh
 ich enables a huge step beyond human capacity toward nanofabricating mater
 ials automatically. MENNDL has been scaled to the 4,200 available nodes of
  Summit achieving a measured 152.5 PFlops, with an estimated sustained per
 formance of 167 PFlops when the entire machine is available.
URL:https://sc18.supercomputing.org/presentation/?id=gb103&sess=sess466
END:VEVENT
END:VCALENDAR

