We study the use of deep learning techniques to reconstruct the kinematics of the neutral current deep inelastic scattering (DIS) process in electron–proton collisions. In particular, we use simulated data from the ZEUS experiment at the HERA accelerator facility, and train deep neural networks to reconstruct the kinematic variables Q2 and x. Our approach is based on the information used in the classical construction methods, the measurements of the scattered lepton, and the hadronic final state in the detector, but is enhanced through correlations and patterns revealed with the simulated data sets. We show that, with the appropriate selection of a training set, the neural networks sufficiently surpass all classical reconstruction methods on most of the kinematic range considered. Rapid access to large samples of simulated data and the ability of neural networks to effectively extract information from large data sets, both suggest that deep learning techniques to reconstruct DIS kinematics can serve as a rigorous method to combine and outperform the classical reconstruction methods.
@article{osti_1900604,
author = {Diefenthaler, Markus and Farhat, Abdullah and Verbytskyi, Andrii and Xu, Yuesheng},
title = {Deeply learning deep inelastic scattering kinematics},
annote = {We study the use of deep learning techniques to reconstruct the kinematics of the neutral current deep inelastic scattering (DIS) process in electron–proton collisions. In particular, we use simulated data from the ZEUS experiment at the HERA accelerator facility, and train deep neural networks to reconstruct the kinematic variables Q2 and x. Our approach is based on the information used in the classical construction methods, the measurements of the scattered lepton, and the hadronic final state in the detector, but is enhanced through correlations and patterns revealed with the simulated data sets. We show that, with the appropriate selection of a training set, the neural networks sufficiently surpass all classical reconstruction methods on most of the kinematic range considered. Rapid access to large samples of simulated data and the ability of neural networks to effectively extract information from large data sets, both suggest that deep learning techniques to reconstruct DIS kinematics can serve as a rigorous method to combine and outperform the classical reconstruction methods.},
doi = {10.1140/epjc/s10052-022-10964-z},
url = {https://www.osti.gov/biblio/1900604},
journal = {European Physical Journal. C, Particles and Fields (Online)},
issn = {ISSN 1434-6052},
number = {11},
volume = {82},
place = {United States},
publisher = {Springer Nature},
year = {2022},
month = {11}}
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 361, Issue 1-2https://doi.org/10.1016/0168-9002(95)00173-5
Abramowicz, Halina; Caldwell, Allen; Sinkus, Ralph
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 365, Issue 2-3https://doi.org/10.1016/0168-9002(95)00612-5
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 506, Issue 3https://doi.org/10.1016/S0168-9002(03)01368-8
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 386, Issue 2-3https://doi.org/10.1016/S0168-9002(96)00893-5
Journal Article
·
Sun Dec 12 23:00:00 EST 2021
· Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment
·OSTI ID:1871017