A Multi-Stage Machine Learning Approach to Cosmic Ray Detection in the RNO-G Experiment

Agarwal S, Aguilar JA, Alden N, Ali S, Allison P, Betts M, Besson D, Bishop A, Botner O, Bouma S, Buitink S, Camphyn R, Chan J, Chiche S, Clark BA, Coleman A, Couberly K, de Kockere S, de Vries KD, Deaconu C, Giri P, Glaser C, Glüsenkamp T, Gui H, Hallgren A, Hallmann S, Hanson JC, Helbing K, Hendricks B, Henrichs J, Heyer N, Hornhuber C, Huesca Santiago E, Hughes K, Jaitly A, Karg T, Karle A, Kelley JL, Kopper C, Korntheuer M, Kowalski M, Kravchenko I, Krebs R, Kugelmeier M, Lahmann R, Liu CH, Marsee MJ, Mulrey K, Muzio M, Nelles A, Novikov A, Nozdrina A, Oberla E, Oeyen B, Punsuebsay N, Pyras L, Ravn M, Rifaie A, Ryckbosch D, Schlüter F, Scholten O, Seckel D, Seikh MF, Selcuk ZS, Stachurska J, Stoffels J, Toscano S, Tosi D, Tutt J, Van Den Broeck DJ, van Eijndhoven N, Vieregg AG, Vijai A, Welling C, Williams DR, Windischhofer P, Wissel S, Young R, Zink A (2025)

Publication Type: Conference contribution

Publication year: 2025

Journal

Publisher: Sissa Medialab Srl

Book Volume: 501

Conference Proceedings Title: Proceedings of Science

Event location: Genf

DOI: 10.22323/1.501.0319

Abstract

The Radio Neutrino Observatory-Greenland (RNO-G), deployed at Summit Station, Greenland, aims to detect ultra-high-energy (UHE) neutrinos. To maximize sensitivity, RNO-G operates with low trigger thresholds, leading to data dominated by background noise, including thermal and anthropogenic sources. A potential additional background source is associated with cosmic ray signals coming from cosmic ray-induced air-showers and further sub-showers produced in the ice, which can mimic neutrino signals. Understanding these events is crucial for improving event classification in future neutrino searches. To address this, two parallel approaches are being developed within the collaboration. The primary method employs a linear discriminant analysis, while an exploratory approach, which this presentation is focused on, utilizes a three-stage event filtering scheme. This scheme sequentially applies a cut-based thermal noise filter, followed by machine learning classifiers—a boosted decision tree (BDT) and a convolutional neural network (CNN)—trained on both simulated cosmic-ray candidates and real background-dominated data. The method effectively rejects background while preserving high signal efficiency. The presentation showcases this powerful machine learning based analysis, highlighting its performance in distinguishing deep cosmic-ray candidates from other types of impulsive background. These results will inform future RNO-G neutrino searches, enhancing its capability to isolate astrophysical neutrino events.

Authors with CRIS profile

Involved external institutions

Bergische Universität Wuppertal Germany (DE) University of Nebraska-Lincoln United States (USA) (US) Pennsylvania State University (Penn State) United States (USA) (US) Université libre de Bruxelles (ULB) / Free University of Brussels Belgium (BE) University of Chicago United States (USA) (US) The University of Alabama United States (USA) (US) University of Kansas (KU) United States (USA) (US) Ohio State University United States (USA) (US) Wisconsin IceCube Particle Astrophysics Center (WIPAC) United States (USA) (US) Uppsala University Sweden (SE) Vrije Universiteit Brussel (VUB) Belgium (BE) University of Maryland, College Park (UMD) United States (USA) (US) Whittier College United States (USA) (US) Deutsches Elektronen-Synchrotron DESY Germany (DE) Radboud University Nijmegen Netherlands (NL) University of Delaware (UDEL) United States (USA) (US) Universiteit Gent (UGent) / Ghent University Belgium (BE)

How to cite

APA:

Agarwal, S., Aguilar, J.A., Alden, N., Ali, S., Allison, P., Betts, M.,... Zink, A. (2025). A Multi-Stage Machine Learning Approach to Cosmic Ray Detection in the RNO-G Experiment. In Proceedings of Science. Genf, CH: Sissa Medialab Srl.

MLA:

Agarwal, S., et al. "A Multi-Stage Machine Learning Approach to Cosmic Ray Detection in the RNO-G Experiment." Proceedings of the 39th International Cosmic Ray Conference, ICRC 2025, Genf Sissa Medialab Srl, 2025.

BibTeX: Download