BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160544Z
LOCATION:Track 6
DTSTART;TZID=America/New_York:20201112T121000
DTEND;TZID=America/New_York:20201112T125000
UID:submissions.supercomputing.org_SC20_sess212_ws_llvmf106@linklings.com
SUMMARY:Autotuning Search Space for Loop Transformations
DESCRIPTION:Workshop\n\nAutotuning Search Space for Loop Transformations\n
 \nKruse, Wu, Finkel\n\nOne of the challenges for optimizing compilers is t
 o predict whether applying an optimization will improve its execution spee
 d. Programmers may override the compiler's profitability heuristic using o
 ptimization directives such as pragmas in the source code. Machine learnin
 g in the form of autotuning can assist users in finding the best optimizat
 ions for each platform.\n\nIn this paper we propose a loop transformation 
 search space that takes the form of a tree, in contrast to previous approa
 ches that usually use vector spaces to represent loop optimization configu
 rations. We implemented a simple autotuner exploring the search space and 
 applied it to a selected set of PolyBench kernels. While the autotuner is 
 capable of representing every possible sequence of loop transformations an
 d their relations, the results motivate the use of better search strategie
 s such as Monte Carlo tree search to find sophisticated loop transformatio
 ns such as multilevel tiling.\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR