BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Introduction to AI Training Workflows for HPC Support Teams
DTSTART:20261117T080000Z
DTEND:20261118T160000Z
DTSTAMP:20261030T113600Z
UID:indico-event-604@events.hpc-portal.eu
CONTACT:e.ewert@fz-juelich.de
DESCRIPTION:This course takes place on-site at JSC. Online participation i
 s not possible. The course will be held in English.\nCourse Content:\n\nIn
 troduction to deep learning: key concepts of modern AI training workflows\
 nBasics of PyTorch and implementing an AI training workflow\nSetting up a 
 virtual environment and running training workloads on HPC systems\nMonitor
 ing training using TensorBoard and Weights & Biases\nParallelisation strat
 egies for distributed training\nProfiling of training code\, identifying b
 ottlenecks and improving performance\n\n \nAfter the course\, participant
 s will be able to:\n\nExplain the basic concepts and terminology of deep l
 earning\nUnderstand and adapt an AI training workflow in PyTorch\nHelp use
 rs run AI training workflows on HPC systems using suitable parallelisation
  strategies\nUse NVIDIA Nsight Systems to identify performance bottlenecks
  in training and advise users on potential improvements\n\n \nTraining me
 thods used in this course:\n\nPresentations introducing the concepts\nHand
 s-on sessions on the JURECA supercomputer\, including running deep learnin
 g workloads\, collecting profiling traces and analysing the results\nQ&A s
 essions and discussion\n\n \n\n\n\n\nContents level\n\n\nin hours\n\n\nin
  %\n\n\n\n\n\n\nBeginner's contents:\n\n\n3 h\n\n\n21 %\n\n\n\n\nIntermedi
 ate contents:\n\n\n6 h\n\n\n43 %\n\n\n\n\nAdvanced contents:\n\n\n5 h\n\n\
 n36 %\n\n\n\n\nCommunity-targeted contents:\n\n\n0 h\n\n\n0 %\n\n\n\n\n \
 nPrerequisites:\nParticipants should have experience with HPC systems and 
 bring their own laptop.\nA personal institutional email address (universit
 y/research institution\, government agency\, organisation\, or company) is
  required to register for JSC training courses. If you don't have an insti
 tutional email address\, please get in touch with the contact person for t
 his course.\nTarget Audience:\nThe course is primarily aimed at HPC suppor
 t staff who would like to help users run and optimise machine learning wor
 kloads. It is also suitable for participants with HPC experience intereste
 d in implementing and profiling their own training workflows.\nLanguage:\n
 The course is given in English.\nDuration:\n2 days\nDates:\n17.-18. Novemb
 er 2026\, 09:00 - 12:00 and 13:00-17:00 each day\nVenue:\nJülich Supercom
 puting Centre\, Building 16.4\, Rotunda\nNumber of Participants:\nMinimum 
 5\, maximum 30 participants\nInstructors:\nStefan Kesselheim\, Sabrina Ben
 assou\, Javad Kasravi\, Eske Ewert\n\nhttps://events.hpc-portal.eu/event/6
 04/
LOCATION:Rotunda (Jülich Supercomputing Centre\, Building 16.4)
URL:https://events.hpc-portal.eu/event/604/
END:VEVENT
END:VCALENDAR