Fri, 04 Oct
|Track Room 4
Systems for Cost-Efficient AI
This workshop addresses the challenges and opportunities in designing software infrastructure for training and serving large-scale AI models, highlighting advances in cost-efficient AI training on public clouds and GPU clusters, & discussing future challenges for foundation and generative AI models
Time & Location
04 Oct 2024, 09:30
Track Room 4, Elias-Canetti-Strasse 146, 8050 Zürich, Switzerland
About the event
Host: Prof. Ana Klimovic
Co-Organizers: Xiaozhe Yao, Maximilian Böther, and Foteini Strati
Keynote speakers: Deepak Narayanan (NVIDIA) and Peter Pietzuch (Imperial College London)
This workshop explores the challenges and opportunities in designing software system infrastructure for training and serving large-scale AI models.
Learning objectives:
Learn about advances in software infrastructure for fast, cost-efficient AI training and serving on the public cloud or in on-premise GPU clusters.
Schedule
45 minutesKeynote
10 minutesModyn: Data-Centric ML Pipeline Orchestration