Compute Accelerator Forum - CADNA, numerical stability; AI-augmented facilities

Wednesday 26 Apr 2023, 16:30 → 18:00 Europe/Zurich

31/S-028 (CERN)

 

To receive annuoncements and information about this forum please subscribe to compute-accelerator-forum-announce@cern.ch

 

Video - CDS

Video - YouTube

1 News

Speakers: Benjamin Morgan (University of Warwick (GB)), Graeme A Stewart (CERN), Dr Maria Girone (CERN), Michael Bussmann, Stefan Roiser (CERN)

Compute Accelerator Forum Introduction, 2023-04-26.pdf

2 CADNA a tool for estimating numerical stability

Speaker: Fabienne Jézéquel (LIP6, Sorbonne University)

slides_FJezequel_CERN_2023.pdf

3 The case for AI-augmented facilities

AI complements human abilities in a way previous ways of computerized automation could not. While in simulation, data analytics and control the use of human written algorithms has achieved tremendous success, AI-driven methods enter areas hitherto though to be part of the human domain. From finding new patterns in data to modeling complex systems and allowing for conversational access to compute resources, AI can add to the mix of human-in-the-loop science at research facilities. We argue that this addition can help to make facilities cater better to the needs of scientific communities and single researchers alike, without relying on the persuasive allure of premature automation based on black box approaches.

Speaker: Dr Michael Bussmann

230426_cern_aiaugmentedfacilities.pptx