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:20260422T143141Z
LOCATION:B302
DTSTART;TZID=America/New_York:20241118T141500
DTEND;TZID=America/New_York:20241118T142000
UID:submissions.supercomputing.org_SC24_sess756_ws_worksa104@linklings.com
SUMMARY:Integrating Evolutionary Algorithms with Distributed Deep Learning
  for Optimizing Hyperparameters on HPC Systems
DESCRIPTION:Mark Coletti, Renan Santos Souza, Tyler Skluzacek, Frédéric Su
 ter, and Rafael Ferreira da Silva (Oak Ridge National Laboratory (ORNL))\n
 \nHigh performance computing (HPC) systems have become essential for solvi
 ng complex scientific problems, particularly in the context of deep learni
 ng (DL). This extended abstract presents a novel system that uses a multio
 bjective evolutionary algorithm (EA) to optimize hyperparameters for a dee
 p learning model, AtomAI, to minimize validation training loss and energy 
 use.  We will be using the parallel and distributed computing capabilities
  of Dask and the scalable provenance features of FlowCept to measure CPU a
 nd \nGPU resource usage as proxies for energy consumption. Our approach fo
 cuses on integrating multiple software components to operate efficiently o
 n large-scale HPC systems, specifically targeting the OLCF's Frontier supe
 rcomputer, but should be generalizable to other HPC environments.\n\nTag: 
 Applications and Application Frameworks, Distributed Computing, Middleware
  and System Software\n\nRegistration Category: Workshop Reg Pass\n\nSessio
 n Chairs: Silvina Caino-Lores (National Institute for Research in Digital 
 Science and Technology (Inria)) and Anirban Mandal\n\n
END:VEVENT
END:VCALENDAR