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:20211207T055352Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20211117T083000
DTEND;TZID=America/Chicago:20211117T170000
UID:submissions.supercomputing.org_SC21_sess279_rpost163@linklings.com
SUMMARY:Code Generation and Optimization for Deep-Learning Computations on
  GPUs via Multi-Dimensional Homomorphisms
DESCRIPTION:Posters, Research Posters\n\nCode Generation and Optimization 
 for Deep-Learning Computations on GPUs via Multi-Dimensional Homomorphisms
 \n\nSchulze, Rasch, Gorlatch\n\nWe present our work-in-progress code gener
 ation and optimization approach for DL computations based on the algebraic
  formalism of multi-dimensional homomorphisms (MDH). We show that popular 
 DL computations can be expressed in the MDH formalism, thereby exploiting 
 the already existing MDH GPU code generation and optimization approach whi
 ch so far has not been focused on DL. Furthermore, we show that the MDH fo
 rmalism is more expressive than the state-of-the-art DL abstractions (e.g.
 , as provided by TensorFlow): for example, MDH can express multiple DL com
 putations (e.g., multiple element-wise computations) as a single MDH expre
 ssion, enabling MDH optimizations (like tiling and parallelization) across
  the computations. Our experiments confirm that our MDH-based approach ach
 ieves better performance than the state-of-the-art, including Apache TVM a
 nd Facebook’s TC.\n\nRegistration Category: Tech Program Reg Pass, Exhibit
  Hall Only
END:VEVENT
END:VCALENDAR