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:C140/142
DTSTART;TZID=America/Chicago:20181115T140000
DTEND;TZID=America/Chicago:20181115T143000
UID:submissions.supercomputing.org_SC18_sess190_pap429@linklings.com
SUMMARY:CosmoFlow: Using Deep Learning to Learn the Universe at Scale
DESCRIPTION:Paper\nApplications, Cosmology, Data Analytics, Deep Learning,
  Machine Learning, Programming Systems, Storage, Visualization, Tech Progr
 am Reg Pass\n\nCosmoFlow: Using Deep Learning to Learn the Universe at Sca
 le\n\nMathuriya, Bard, Mendygral, Meadows, Arnemann...\n\nDeep learning is
  a promising tool to determine the physical model that describes our unive
 rse.   To handle the considerable computational cost of this problem, we p
 resent CosmoFlow: a highly scalable deep learning application built on top
  of the TensorFlow framework.\n\nCosmoFlow uses efficient implementations 
 of 3D convolution and pooling primitives, together with improvements in th
 reading for many element-wise operations, to improve training performance 
 on Intel Xeon Phi processors.  We also utilize the Cray PE Machine Learnin
 g Plugin for efficient scaling to multiple nodes. We demonstrate fully syn
 chronous data-parallel training on 8192 nodes of Cori with 77% parallel ef
 ficiency, achieving 3.5 Pflop/s sustained performance. \n\nTo our knowledg
 e, this is the first large-scale science application of the TensorFlow fra
 mework at supercomputer scale with fully-synchronous training. These enhan
 cements enable us to process large 3D dark matter distribution and predict
  the cosmological parameters Omega_M, sigma_8 and N_s with unprecedented a
 ccuracy.
URL:https://sc18.supercomputing.org/presentation/?id=pap429&sess=sess190
END:VEVENT
END:VCALENDAR