Identification of heavy, energetic, hadronically decaying particles using machine-learning techniques

Identification of heavy, energetic, hadronically decaying particles using machine-learning techniques

Մատենագիտական մանրամասներ
Parent link:Journal of Instrumentation
Vol. 15, iss. 6.— 2020.— [P06005, 88 p.]
Համատեղ հեղինակ: Национальный исследовательский Томский политехнический университет Исследовательская школа физики высокоэнергетических процессов
Այլ հեղինակներ: Sirunyan A. M., Tumasyan A. R., Adam W. Wolfgang, Ambrogi F. Federico, Bergauer T. Thomas, Babaev A. A. Anton Anatoljevich, Okhotnikov V. V. Vitaly Vladimirovich, Yuzhakov M. M. Mikhail Mikhaylovich
Ամփոփում:Title screen
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 vs = 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.
Հրապարակվել է: 2020
Խորագրեր:
электронный ресурс
труды учёных ТПУ
large detector-systems performance
pattern recognition
cluster finding
calibration
fitting methods
детекторные устройства
распознавание образов
кластеры
калибровка
Առցանց հասանելիություն:https://doi.org/10.1088/1748-0221/15/06/P06005
Ձևաչափ: Էլեկտրոնային Գրքի գլուխ
KOHA link:https://koha.lib.tpu.ru/cgi-bin/koha/opac-detail.pl?biblionumber=662526
Նկարագրություն
Ամփոփում:Title screen
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 vs = 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.
DOI:10.1088/1748-0221/15/06/P06005

Նմանատիպ նյութեր