BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20190719T085743Z
LOCATION:HG F 1
DTSTART;TZID=Europe/Stockholm:20190612T140000
DTEND;TZID=Europe/Stockholm:20190612T143000
UID:submissions.pasc-conference.org_PASC19_sess138_msa264@linklings.com
SUMMARY:Sparse HPC Opportunities in Deep Neural Networks
DESCRIPTION:Minisymposium\nComputer Science and Applied Mathematics, Chemi
 stry and Materials, Physics, Engineering\n\nSparse HPC Opportunities in De
 ep Neural Networks\n\nKoanantakool\n\nWith the wide adoption of deep learn
 ing, there have been numerous efforts to accelerate deep neural networks, 
 one of which is to exploit sparsity. Many high-dimensional problems are sp
 arse by nature, and researchers have come up with abundant ways to enforce
  sparse connections. In this talk, we give examples of where sparsity
  may arise in deep neural networks, both naturally (e.g., sparse inputs) a
 nd artificially (e.g., sparse weights and filters – unstructured or 
 structured). We also discuss how the current machine learning frameworks h
 andle sparse computations and the opportunities to apply optimization tech
 niques from high-performance computing.
END:VEVENT
END:VCALENDAR