15–19 Sept 2025
CERN
Europe/Zurich timezone

Optimized and scalable statistical analysis and numerical optimization with autograd and ML frameworks

15 Sept 2025, 14:35
5m
500/1-001 - Main Auditorium (CERN)

500/1-001 - Main Auditorium

CERN

400
Show room on map
1. Cutting Edge AI for Offline Data Processing Cutting Edge AI for Offline Data Processing

Speaker

Josh Bendavid (CERN)

Description

High performance maximum likelihood fitting and associated statistical analysis tools, initially based on Tensorflow 1 (combinetf) and used for precision measurements (CMS W helicity, W mass), now re-written with Tensorflow 2 (RABBIT) and being used for in-progress CMS alphaS measurement, etc. Older combinetf has also been used for some of the analyses in the FCC feasibility study

CERN group/ Experiment

EP-CMG, EP-FCC

Working area Area 1" Cutting Edge AI for Offline Data Processing
Project goals Continue developing functionality and facilitating general use of Rabbit for LHC and FCC analysis
Timeline Additional features, documentation, and support over the next 3 years.
Available person power 10-20% of two postdocs
Additional person power request Supervision from staff
Is this an already ongoing activity? Yes
Indicative hardware resources needs Access to a GPU and a CPU cluster or individual machines with many cores, with LCG-like software stack and cvmfs access with fast storage facilities across the full duration of the project

Author

Co-authors

David Walter (Massachusetts Inst. of Technology (US)) Jan Eysermans (Massachusetts Inst. of Technology (US))

Presentation materials

There are no materials yet.