2020/08/20 by Mohan, Lakshan Ram Madhan, Marshall, Alexander, Maddrell-Mander, Samuel +5
#Computational Physics (physics.comp-ph) #FOS: Physical sciences #High Energy Physics - Experiment (hep-ex)
paper · doi:10.48550/arxiv.2008.09210
This paper presents the first study of Graphcore's Intelligence Processing Unit (IPU) in the context of particle physics applications. The IPU is a new type of processor optimised for machine learning. Comparisons are made for neural-network-based event simulation, multiple-scattering correction, and flavour tagging, implemented on IPUs, GPUs and CPUs, using a variety of neural network architectures and hyperparameters. Additionally, a Kálmán filter for track reconstruction is implemented on IPUs and GPUs. The results indicate that IPUs hold considerable promise in addressing the rapidly increasing compute needs in particle physics.