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:20181221T160726Z
LOCATION:D166
DTSTART;TZID=America/Chicago:20181111T140000
DTEND;TZID=America/Chicago:20181111T143000
UID:submissions.supercomputing.org_SC18_sess174_ws_exampi104@linklings.com
SUMMARY:AITuning: Machine Learning-Based Tuning Tool for Run-Time Communic
 ation Libraries
DESCRIPTION:Workshop\nExascale, MPI, Networks, System Software, Workshop R
 eg Pass\n\nAITuning: Machine Learning-Based Tuning Tool for Run-Time Commu
 nication Libraries\n\nFanfarillo, Del Vento\n\nIn this work, we address th
 e problem of tuning communication libraries by using a deep reinforcement 
 learning approach.  Reinforcement learning is a machine learning technique
  incredibly effective in solving game-like situations.  In fact, tuning a 
 set of parameters in a communication library in order to get better perfor
 mance in a parallel application can be expressed as a game: find the right
  combination/path that provides the best reward.  Even though AITuning has
  been designed to be utilized with different run-time libraries, we focuse
 d this work on applying it to the OpenCoarrays run-time communication libr
 ary, built on top of MPI-3.  This work not only shows the potential of usi
 ng a reinforcement learning algorithm for tuning communication libraries, 
 but also demonstrates how the MPI Tool Information Interface, introduced b
 y the MPI-3 standard, can be used effectively by run-time libraries to imp
 rove the performance without human intervention.
URL:https://sc18.supercomputing.org/presentation/?id=ws_exampi104&sess=ses
 s174
END:VEVENT
END:VCALENDAR