2015/01/23 by Leandro G. Almeida, Mihailo Backović, Almeida, Leandro G. +7 · 3 citations
Computer Science · Physics and Astronomy · #Computational Physics and Python Applications #FOS: Physical sciences #High Energy Physics - Experiment (hep-ex) #High Energy Physics - Phenomenology (hep-ph) #High-Energy Particle Collisions Research #Particle physics theoretical and experimental studies
paper · pdf · doi:10.48550/arxiv.1501.05968
openalex publication_date 2015/01/23 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
Many searches for physics beyond the Standard Model at the Large Hadron\nCollider (LHC) rely on top tagging algorithms, which discriminate between\nboosted hadronic top quarks and the much more common jets initiated by light\nquarks and gluons. We note that the hadronic calorimeter (HCAL) effectively\ntakes a "digital image" of each jet, with pixel intensities given by energy\ndeposits in individual HCAL cells. Viewed in this way, top tagging becomes a\ncanonical pattern recognition problem. With this motivation, we present a novel\ntop tagging algorithm based on an Artificial Neural Network (ANN), one of the\nmost popular approaches to pattern recognition. The ANN is trained on a large\nsample of boosted tops and light quark/gluon jets, and is then applied to\nindependent test samples. The ANN tagger demonstrated excellent performance in\na Monte Carlo study: for example, for jets with pT in the 1100-1200 GeV range,\n60% top-tag efficiency can be achieved with a 4% mis-tag rate. We discuss the\nphysical features of the jets identified by the ANN tagger as the most\nimportant for classification, as well as correlations between the ANN tagger\nand some of the familiar top-tagging observables and algorithms.\n