BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Multi-GPU optimisation of hyperparameters and AI models on a super
 computer
DTSTART:20261127T080000Z
DTEND:20261127T120000Z
DTSTAMP:20261030T093600Z
UID:indico-event-603@events.hpc-portal.eu
DESCRIPTION:\nDo you train artificial intelligence models and want to impr
 ove their performance without spending hours manually testing different co
 nfigurations?\nThis training course will teach you how to optimize the hyp
 erparameters of your AI models and harness the power of GPUs and a superco
 mputer’s multi-GPU environments.\n \nCourse outline:\n\nUnderstand hype
 rparameters and the main methods for optimizing them\nDiscover the concept
  of pruning\nOptimize hyperparameters on a GPU\nScale up in a multi-GPU en
 vironment\nAnalyze model performance\nCompare and discuss the gains achiev
 ed after optimization\nPut these concepts into practice through exercises 
 on the ROMEO2025 supercomputer’s GPUs\n\nWhy participate?\n\nLearn how t
 o optimize the hyperparameters of your artificial intelligence models\nUnd
 erstand the metrics used to evaluate a model’s performance\nReduce the t
 ime required to find the best configurations\nEffectively leverage multi-G
 PU resources\nConduct experiments directly on the ROMEO2025 supercomputer\
 n\nTarget Audience:\nThis training is intended for engineers and researche
 rs who wish to optimize their models and hyperparameters by leveraging the
  multi-GPU computing power available on the nodes of a supercomputer such 
 as ROMEO.\n\n \n\n\nhttps://events.hpc-portal.eu/event/603/
URL:https://events.hpc-portal.eu/event/603/
END:VEVENT
END:VCALENDAR