2018/10/18 by Jesús Arjona Martínez, Martinez, Jesus Arjona, Olmo Cerri +7
Engineering · Physics and Astronomy · #FOS: Computer and information sciences #FOS: Physical sciences #High Energy Physics - Experiment (hep-ex) #High Energy Physics - Phenomenology (hep-ph) #High-Energy Particle Collisions Research #Machine Learning (cs.LG) #Particle Detector Development and Performance #Particle physics theoretical and experimental studies #Superconducting Materials and Applications
paper · pdf · doi:10.48550/arxiv.1810.07988
openalex publication_date 2018/10/18 · openalex created_date 2022/08/02 · openalex updated_date 2026/07/28
At the Large Hadron Collider, the high transverse-momentum events studied by\nexperimental collaborations occur in coincidence with parasitic low\ntransverse-momentum collisions, usually referred to as pileup. Pileup\nmitigation is a key ingredient of the online and offline event reconstruction\nas pileup affects the reconstruction accuracy of many physics observables. We\npresent a classifier based on Graph Neural Networks, trained to retain\nparticles coming from high-transverse-momentum collisions, while rejecting\nthose coming from pileup collisions. This model is designed as a refinement of\nthe PUPPI algorithm, employed in many LHC data analyses since 2015. Thanks to\nan extended basis of input information and the learning capabilities of the\nconsidered network architecture, we show an improvement in pileup-rejection\nperformances with respect to state-of-the-art solutions.\n