10-15 March 2019
Steinmatte conference center
Europe/Zurich timezone

Pileup mitigation with graph neural networks

Not scheduled
Steinmatte conference center

Steinmatte conference center

Hotel Allalin, Saas Fee, Switzerland https://allalin.ch/conference/
Poster Track 2: Data Analysis - Algorithms and Tools Poster Session


Olmo Cerri (California Institute of Technology (US))


Mitigation of the effect of the multiple parasitic proton collisions produced during bunch crossing at the LHC is a major endeavor towards the realization of the physics program at the collider. The pileup affects many physics observable derived during the online and offline reconstruction. We propose a graph neural network machine learning model, based on the PUPPI approach, for identifying particle coming from pileup and retaining the ones from high-transverse momentum collisions. We show improvement in pileup rejection performance and energy resolution with respect to solutions currently used at the LHC.

Primary authors

Dr Jean-Roch Vlimant (California Institute of Technology (US)) Maurizio Pierini (CERN) Olmo Cerri (California Institute of Technology (US)) Jesus Arjona Martinez (California Institute of Technology (US)) Maria Spiropulu (California Institute of Technology)

Presentation materials