15–19 Sept 2025
CERN
Europe/Zurich timezone

federated learning for cross-experiment foundation models

15 Sept 2025, 10:40
5m
500/1-001 - Main Auditorium (CERN)

500/1-001 - Main Auditorium

CERN

400
Show room on map
4. AI Infrastructure for Model Training Cutting Edge AI for Offline Data Processing

Speaker

Maurizio Pierini (CERN)

Description

Several reconstruction steps in LHC events are being approached using end-to-end deep learning solutions (e.g., for tracking, calorimetry, and particle flow linking). It has been proposed that foundation models trained on physics events could repeat for LHC event reconstruction the astonishing success of Large Language Models in developing multitasking skills. Such an application could be trained on particle flow reconstruction, starting from raw detector inputs to build a global event view without training supervision (unlike the state-of-the-art MLPF model). First efforts towards this Large Physics Models have shown that encouraging results. The use of federated learning could facilitate these efforts: by simultaneously extracting knowledge from various experimental datasets, one could abstract each detector data to a common space, where the underlying physics could be learned more accurately. The knowledge gained with reconstructing CMS data could improve the ATLAS reconstruction, and viceversa. Federated the experience of the CAFEIN team at CERN on large-scale federated learning, we propose to establish an effort to experiment on the idea of cross-experiment learning and assess its potential to improve the physics performance of the various experiments. The success of this process would offer a platform to transfer knowledge from the LHC experiments to future colliders.

CERN group/ Experiment

EP and ATS

Working area Area 4: AI Infrastructure for Model Training
Project goals develop a demonstrator of federative learning capabilities for cross-experiment applications
Timeline 3 years
Available person power 0
Additional person power request 2 PhD
Is this an already ongoing activity? No
Indicative hardware resources needs large cluster of H100/A100 GPUs

Authors

Presentation materials

There are no materials yet.