Identification of proton and gamma in LHAASO-KM2A simulation data with deep learning algorithms
Issued Date
2022-03-18
Resource Type
eISSN
18248039
Scopus ID
2-s2.0-85123976685
Journal Title
Proceedings of Science
Volume
395
Rights Holder(s)
SCOPUS
Bibliographic Citation
Proceedings of Science Vol.395 (2022)
Suggested Citation
Zhang F. Identification of proton and gamma in LHAASO-KM2A simulation data with deep learning algorithms. Proceedings of Science Vol.395 (2022). Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/86591
Title
Identification of proton and gamma in LHAASO-KM2A simulation data with deep learning algorithms
Author(s)
Author's Affiliation
State Key Laboratory of Particle Detection & Electronics
Nanjing University
Shanghai Astronomical Observatory Chinese Academy of Sciences
Shandong University
Wuhan University
Yunnan University
Institute of High Energy Physics Chinese Academy of Science
University of Chinese Academy of Sciences
Guangzhou University
Tsinghua University
Sun Yat-Sen University
University of Science and Technology of China
Zhengzhou University
Institiúid Ard-Lénn Bhaile Átha Cliath
Università degli Studi di Napoli Federico II
Sichuan University
National Astronomical Observatories Chinese Academy of Sciences
Max-Planck-Institut für Kernphysik
Southwest Jiaotong University
Purple Mountain Observatory Chinese Academy of Sciences
Hebei Normal University
Tibet University
Universit'e de Gen'eve
TIANFU Cosmic Ray Research Center
Nanjing University
Shanghai Astronomical Observatory Chinese Academy of Sciences
Shandong University
Wuhan University
Yunnan University
Institute of High Energy Physics Chinese Academy of Science
University of Chinese Academy of Sciences
Guangzhou University
Tsinghua University
Sun Yat-Sen University
University of Science and Technology of China
Zhengzhou University
Institiúid Ard-Lénn Bhaile Átha Cliath
Università degli Studi di Napoli Federico II
Sichuan University
National Astronomical Observatories Chinese Academy of Sciences
Max-Planck-Institut für Kernphysik
Southwest Jiaotong University
Purple Mountain Observatory Chinese Academy of Sciences
Hebei Normal University
Tibet University
Universit'e de Gen'eve
TIANFU Cosmic Ray Research Center
Other Contributor(s)
Abstract
Identification of proton and gamma plays an essential role in ultra-high energy gamma-ray astronomy with LHAASO-KM2A. In this work, two neural networks (deep neural networks (DNN) and graph neural networks (GNN)) are applied to distinguish proton and gamma in the LHAASOKM2A simulation data. The receiver operating characteristic (ROC) curves are used to evaluate the quality of the model. Both KM2A-DNN and KM2A-GNN models give higher Area Under Curve (AUC) scores than the traditional baseline model.