par Sirunyan, A.M.;Beghin, Diego ;Bilin, Bugra ;Clerbaux, Barbara ;De Lentdecker, Gilles ;Delannoy, Hugo ;Dorney, Brian ;Favart, Laurent ;Grebenyuk, Anastasia ;Kalsi, Amandeep Kaur ;Moureaux, Louis ;Popov, Andrey ;Postiau, Nicolas ;Starling, Elizabeth Rose ;Thomas, Laurent ;Vander Velde, Catherine ;Vannerom, David ;Malara, Andrea ; [et al.]
Référence Journal of Instrumentation, 15, 6, P06005
Publication Publié, 2020-06-01
Référence Journal of Instrumentation, 15, 6, P06005
Publication Publié, 2020-06-01
Article révisé par les pairs
Résumé : | Machine-learning (ML) techniques are explored to identify and classify hadronic decays of highly Lorentz-boosted W/Z/Higgs bosons and top quarks. Techniques without ML have also been evaluated and are included for comparison. The identification performances of a variety of algorithms are characterized in simulated events and directly compared with data. The algorithms are validated using proton-proton collision data at s = 13TeV, corresponding to an integrated luminosity of 35.9 fb-1. Systematic uncertainties are assessed by comparing the results obtained using simulation and collision data. The new techniques studied in this paper provide significant performance improvements over non-ML techniques, reducing the background rate by up to an order of magnitude at the same signal efficiency. |