Standard

Towards mass composition study with KASCADE using deep neural networks. / Kuznetsov, Mikhail; Petrov, Nikita; Plokhikh, Ivan et al.

In: Proceedings of Science, Vol. 423, 092, 14.12.2023.

Research output: Contribution to journalConference articlepeer-review

Harvard

Kuznetsov, M, Petrov, N, Plokhikh, I & Sotnikov, V 2023, 'Towards mass composition study with KASCADE using deep neural networks', Proceedings of Science, vol. 423, 092. https://doi.org/10.22323/1.423.0092

APA

Kuznetsov, M., Petrov, N., Plokhikh, I., & Sotnikov, V. (2023). Towards mass composition study with KASCADE using deep neural networks. Proceedings of Science, 423, [092]. https://doi.org/10.22323/1.423.0092

Vancouver

Kuznetsov M, Petrov N, Plokhikh I, Sotnikov V. Towards mass composition study with KASCADE using deep neural networks. Proceedings of Science. 2023 Dec 14;423:092. doi: 10.22323/1.423.0092

Author

Kuznetsov, Mikhail ; Petrov, Nikita ; Plokhikh, Ivan et al. / Towards mass composition study with KASCADE using deep neural networks. In: Proceedings of Science. 2023 ; Vol. 423.

BibTeX

@article{6380bdc316d74198ad55888e6b43d5e4,
title = "Towards mass composition study with KASCADE using deep neural networks",
abstract = "We present new insight into the ongoing machine learning analysis of KASCADE experiment archival data, that contain air shower events with ∼ 1 − 100 PeV primary energy. The aim of the study is to improve the accuracy of high-energy cosmic rays mass composition reconstruction with respect to the standard KASCADE technique. We introduce five mass groups: protons, helium, carbon, silicon and iron nuclei and interpret the reconstruction process as a classification task. We employ a random forest technique as well as two promising neural network architectures — a self-attention perceptron and a convolutional neural network. These models are being trained with KASCADE CORSIKA simulations. We examine the behavior of the mass composition reconstruction for several hadronic interaction models and additionally check the credibility of our methods with a small {"}unblinded{"} part of the real KASCADE data.",
author = "Mikhail Kuznetsov and Nikita Petrov and Ivan Plokhikh and Vladimir Sotnikov",
note = "The work is supported by the Russian Science Foundation grant 22-22-00883. We appreciate the contribution on the initial stages of this project made by our colleagues: Dmitry Kostunin, Vladimir Lenok and Victoria Tokareva.; 27th European Cosmic Ray Symposium ; Conference date: 25-07-2022 Through 29-07-2022",
year = "2023",
month = dec,
day = "14",
doi = "10.22323/1.423.0092",
language = "English",
volume = "423",
journal = "Proceedings of Science",
issn = "1824-8039",
publisher = "Sissa Medialab Srl",

}

RIS

TY - JOUR

T1 - Towards mass composition study with KASCADE using deep neural networks

AU - Kuznetsov, Mikhail

AU - Petrov, Nikita

AU - Plokhikh, Ivan

AU - Sotnikov, Vladimir

N1 - Conference code: 27

PY - 2023/12/14

Y1 - 2023/12/14

N2 - We present new insight into the ongoing machine learning analysis of KASCADE experiment archival data, that contain air shower events with ∼ 1 − 100 PeV primary energy. The aim of the study is to improve the accuracy of high-energy cosmic rays mass composition reconstruction with respect to the standard KASCADE technique. We introduce five mass groups: protons, helium, carbon, silicon and iron nuclei and interpret the reconstruction process as a classification task. We employ a random forest technique as well as two promising neural network architectures — a self-attention perceptron and a convolutional neural network. These models are being trained with KASCADE CORSIKA simulations. We examine the behavior of the mass composition reconstruction for several hadronic interaction models and additionally check the credibility of our methods with a small "unblinded" part of the real KASCADE data.

AB - We present new insight into the ongoing machine learning analysis of KASCADE experiment archival data, that contain air shower events with ∼ 1 − 100 PeV primary energy. The aim of the study is to improve the accuracy of high-energy cosmic rays mass composition reconstruction with respect to the standard KASCADE technique. We introduce five mass groups: protons, helium, carbon, silicon and iron nuclei and interpret the reconstruction process as a classification task. We employ a random forest technique as well as two promising neural network architectures — a self-attention perceptron and a convolutional neural network. These models are being trained with KASCADE CORSIKA simulations. We examine the behavior of the mass composition reconstruction for several hadronic interaction models and additionally check the credibility of our methods with a small "unblinded" part of the real KASCADE data.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85180311400&origin=inward&txGid=c3aa4e985c331a3096732842e837d744

UR - https://www.mendeley.com/catalogue/5249092d-9515-3c75-9d6f-6fe641ab18d5/

U2 - 10.22323/1.423.0092

DO - 10.22323/1.423.0092

M3 - Conference article

VL - 423

JO - Proceedings of Science

JF - Proceedings of Science

SN - 1824-8039

M1 - 092

T2 - 27th European Cosmic Ray Symposium

Y2 - 25 July 2022 through 29 July 2022

ER -

ID: 59538945