Mass composition of Telescope Array’s surface detectors events using deep learning

Telescope Array Collaboration

Mass composition of Telescope Array’s surface detectors events using deep learning

Telescope Array Collaboration

Department of Physics

Research output: Contribution to journal › Conference article › peer-review

Abstract

We report on an improvement of deep learning techniques used for identifying primary particles of atmospheric air showers. The progress was achieved by using two neural networks. The first works as a classifier for individual events, while the second predicts fractions of elements in an ensemble of events based on the inference of the first network. For a fixed hadronic model, this approach yields an accuracy of 90% in identifying fractions of elements in an ensemble of events.

Original language	English
Article number	384
Journal	Proceedings of Science
Volume	395
Publication status	Published - 2022 Mar 18
Event	37th International Cosmic Ray Conference, ICRC 2021 - Virtual, Berlin, Germany Duration: 2021 Jul 12 → 2021 Jul 23

Bibliographical note

Publisher Copyright:
© Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0).

All Science Journal Classification (ASJC) codes

General

Cite this

@article{045a670c6a804466b39d9ea05e07c69e,

title = "Mass composition of Telescope Array{\textquoteright}s surface detectors events using deep learning",

abstract = "We report on an improvement of deep learning techniques used for identifying primary particles of atmospheric air showers. The progress was achieved by using two neural networks. The first works as a classifier for individual events, while the second predicts fractions of elements in an ensemble of events based on the inference of the first network. For a fixed hadronic model, this approach yields an accuracy of 90% in identifying fractions of elements in an ensemble of events.",

author = "{Telescope Array Collaboration} and I. Kharuk and O. Kalashev and Abbasi, {R. U.} and T. Abu-Zayyad and M. Allen and Y. Arai and R. Arimura and E. Barcikowski and Belz, {J. W.} and Bergman, {D. R.} and Blake, {S. A.} and I. Buckland and R. Cady and Cheon, {B. G.} and J. Chiba and M. Chikawa and T. Fujii and K. Fujisue and K. Fujita and R. Fujiwara and M. Fukushima and R. Fukushima and G. Furlich and R. Gonzalez and W. Hanlon and M. Hayashi and N. Hayashida and K. Hibino and R. Higuchi and K. Honda and D. Ikeda and T. Inadomi and N. Inoue and T. Ishii and H. Ito and D. Ivanov and H. Iwakura and A. Iwasaki and Jeong, {H. M.} and S. Jeong and Jui, {C. C.H.} and K. Kadota and F. Kakimoto and O. Kalashev and K. Kasahara and S. Kasami and H. Kawai and S. Kawakami and S. Kawana and Kwon, {Y. J.}",

note = "Publisher Copyright: {\textcopyright} Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0).; 37th International Cosmic Ray Conference, ICRC 2021 ; Conference date: 12-07-2021 Through 23-07-2021",

year = "2022",

month = mar,

day = "18",

language = "English",

volume = "395",

journal = "Proceedings of Science",

issn = "1824-8039",

publisher = "Sissa Medialab Srl",

}

TY - JOUR

T1 - Mass composition of Telescope Array’s surface detectors events using deep learning

AU - Telescope Array Collaboration

AU - Kharuk, I.

AU - Kalashev, O.

AU - Abbasi, R. U.

AU - Abu-Zayyad, T.

AU - Allen, M.

AU - Arai, Y.

AU - Arimura, R.

AU - Barcikowski, E.

AU - Belz, J. W.

AU - Bergman, D. R.

AU - Blake, S. A.

AU - Buckland, I.

AU - Cady, R.

AU - Cheon, B. G.

AU - Chiba, J.

AU - Chikawa, M.

AU - Fujii, T.

AU - Fujisue, K.

AU - Fujita, K.

AU - Fujiwara, R.

AU - Fukushima, M.

AU - Fukushima, R.

AU - Furlich, G.

AU - Gonzalez, R.

AU - Hanlon, W.

AU - Hayashi, M.

AU - Hayashida, N.

AU - Hibino, K.

AU - Higuchi, R.

AU - Honda, K.

AU - Ikeda, D.

AU - Inadomi, T.

AU - Inoue, N.

AU - Ishii, T.

AU - Ito, H.

AU - Ivanov, D.

AU - Iwakura, H.

AU - Iwasaki, A.

AU - Jeong, H. M.

AU - Jeong, S.

AU - Jui, C. C.H.

AU - Kadota, K.

AU - Kakimoto, F.

AU - Kalashev, O.

AU - Kasahara, K.

AU - Kasami, S.

AU - Kawai, H.

AU - Kawakami, S.

AU - Kawana, S.

AU - Kwon, Y. J.

N1 - Publisher Copyright: © Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0).

PY - 2022/3/18

Y1 - 2022/3/18

N2 - We report on an improvement of deep learning techniques used for identifying primary particles of atmospheric air showers. The progress was achieved by using two neural networks. The first works as a classifier for individual events, while the second predicts fractions of elements in an ensemble of events based on the inference of the first network. For a fixed hadronic model, this approach yields an accuracy of 90% in identifying fractions of elements in an ensemble of events.

AB - We report on an improvement of deep learning techniques used for identifying primary particles of atmospheric air showers. The progress was achieved by using two neural networks. The first works as a classifier for individual events, while the second predicts fractions of elements in an ensemble of events based on the inference of the first network. For a fixed hadronic model, this approach yields an accuracy of 90% in identifying fractions of elements in an ensemble of events.

UR - http://www.scopus.com/inward/record.url?scp=85143762556&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85143762556&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85143762556

SN - 1824-8039

VL - 395

JO - Proceedings of Science

JF - Proceedings of Science

M1 - 384

T2 - 37th International Cosmic Ray Conference, ICRC 2021

Y2 - 12 July 2021 through 23 July 2021

ER -

Mass composition of Telescope Array’s surface detectors events using deep learning

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this