Final published version
Licence: CC BY: Creative Commons Attribution 4.0 International License
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - Simultaneous energy and mass calibration of large-radius jets with the ATLAS detector using a deep neural network
AU - The ATLAS collaboration
AU - Barton, A.E.
AU - Borissov, G.
AU - Bouhova-Thacker, E.V.
AU - Ferguson, Ruby
AU - Ferrando, James
AU - Fox, H.
AU - Hagan, Alina
AU - Henderson, R.C.W.
AU - Jones, R.W.L.
AU - Kartvelishvili, V.
AU - Love, P.A.
AU - Marshall, E.J.
AU - Meng, L.
AU - Muenstermann, D.
AU - Ribaric, N.
AU - Rybacki, K.
AU - Smizanska, M.
AU - Spinali, S.
AU - Wharton, A.M.
PY - 2024/9/3
Y1 - 2024/9/3
N2 - The energy and mass measurements of jets are crucial tasks for the Large Hadron Collider experiments. This paper presents a new calibration method to simultaneously calibrate these quantities for large-radius jets measured with the ATLAS detector using a deep neural network (DNN). To address the specificities of the calibration problem, special loss functions and training procedures are employed, and a complex network architecture, which includes feature annotation and residual connection layers, is used. The DNN-based calibration is compared to the standard numerical approach in an extensive series of tests. The DNN approach is found to perform significantly better in almost all of the tests and over most of the relevant kinematic phase space. In particular, it consistently improves the energy and mass resolutions, with a 30% better energy resolution obtained for transverse momenta pT>500 GeV.
AB - The energy and mass measurements of jets are crucial tasks for the Large Hadron Collider experiments. This paper presents a new calibration method to simultaneously calibrate these quantities for large-radius jets measured with the ATLAS detector using a deep neural network (DNN). To address the specificities of the calibration problem, special loss functions and training procedures are employed, and a complex network architecture, which includes feature annotation and residual connection layers, is used. The DNN-based calibration is compared to the standard numerical approach in an extensive series of tests. The DNN approach is found to perform significantly better in almost all of the tests and over most of the relevant kinematic phase space. In particular, it consistently improves the energy and mass resolutions, with a 30% better energy resolution obtained for transverse momenta pT>500 GeV.
KW - detector
KW - calibrations
KW - CERN jets
KW - ATLAS
U2 - 10.1088/2632-2153/ad611e
DO - 10.1088/2632-2153/ad611e
M3 - Journal article
VL - 5
JO - Machine Learning: Science and Technology
JF - Machine Learning: Science and Technology
SN - 2632-2153
IS - 3
M1 - 035051
ER -