Apr 15 – 18, 2019
Europe/Zurich timezone
There is a live webcast for this event.

Learning representations of irregular particle-detector geometry with distance-weighted graph networks

Apr 17, 2019, 3:30 PM
500/1-001 - Main Auditorium (CERN)

500/1-001 - Main Auditorium


Show room on map


Jan Kieseler (CERN)


We explore the possibility of using graph networks to deal with irregular-geometry detectors when reconstructing particles. Thanks to their representation-learning capabilities, graph networks can exploit the detector granularity, while dealing with the event sparsity and the irregular detector geometry. In this context, we introduce two distance-weighted graph network architectures, the GarNet and the GravNet layers and we apply them to a typical particle reconstruction task. As an example, we consider a high granularity calorimeter, loosely inspired by the endcap calorimeter to be installed in the CMS detector for the High-Luminosity LHC phase. We focus the study on the basis for calorimeter reconstruction, clustering, and provide a quantitative comparison to alternative approaches. The proposed methods outperform previous methods or reach competitive performance while keeping favourable computing-resource consumption. Being geometry agnostic, they can be easily generalized to other use cases and to other detectors, e.g., tracking in silicon detectors.

Preferred contribution length 20 minutes

Primary authors

Jan Kieseler (CERN) Yutaro Iiyama (CERN) Shah Rukh Qasim (SEECS - School of Electrical Engine ering and Computer Science ) Maurizio Pierini (CERN)

Presentation materials