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:20230124T171523Z
LOCATION:D220
DTSTART;TZID=America/Chicago:20221113T093000
DTEND;TZID=America/Chicago:20221113T100000
UID:submissions.supercomputing.org_SC22_sess430_ws_llvmf102@linklings.com
SUMMARY:Reinforcement Learning Assisted Loop Distribution for Locality and
  Vectorization
DESCRIPTION:Workshop\n\nReinforcement Learning Assisted Loop Distribution 
 for Locality and Vectorization\n\nJain, VenkataKeerthy, Aggarwal, Dangeti,
  Das...\n\nWe present a Reinforcement Learning (RL) based approach to effi
 ciently perform loop-distribution with the goals of optimizing for vectori
 zation and locality. We generate the SCC Dependence Graph for each loop of
  the program. Our RL model learns to predict the distribution order of the
  loop by performing a topological walk of graph. The RL-reward is computed
  using instruction cost and number of cache misses. For training purposes,
  we also propose a novel strategy to extend the training set by generating
  new loops.\n\nWe show results on x86 architecture on various benchmarks: 
 TSVC, LLVM-Test-Suite, PolyBench, PolyBenchNN. Our framework achieves an a
 verage improvement of 3.63% on TSVC, 4.61% on LLVM-Test-Suite MicroBenchma
 rks, 1.78% on PolyBench and 1.95% on PolyBenchNN benchmark suites for perf
 ormance, with LLVM-O3 flag as baseline. We also show the improvements on o
 ther performance metrics like Instruction Per Cycle (IPC), Number of loops
  distributed and vectorized, and L1 cache performance.\n\nSession Format: 
 Recorded\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR