Deep Learning Methods for Gamma/Hadron Separation in SWGO

Schneider M, Pirke M, Leitl F, Glombitza J, van Eldik C (2025)

Publication Type: Conference contribution

Publication year: 2025

Journal

Publisher: Sissa Medialab Srl

Book Volume: 501

Conference Proceedings Title: Proceedings of Science

Event location: Geneva

DOI: 10.22323/1.501.0836

Abstract

The Southern Wide-field Gamma-ray Observatory (SWGO) is a planned water Cherenkov-based observatory to be located in Pampa La Bola, Chile, providing continuous, wide-field observations of the gamma-ray sky. SWGO will provide a unique view of the wide southern hemisphere gamma-ray sky, complementing other very-high-energy observatories such as HAWC, CTAO, and LHAASO. A key challenge in ground-based gamma-ray astronomy is an effective gamma/hadron separation to suppress the dominant cosmic-ray background. In this contribution, we will present ongoing studies on advanced classification techniques for SWGO, specifically exploring deep learning approaches based on Graph Neural Networks and Transformer architectures. These methods, currently tested through simulations, offer promising advancements in gamma/hadron separation and event reconstruction.

Authors with CRIS profile

How to cite

APA:

Schneider, M., Pirke, M., Leitl, F., Glombitza, J., & van Eldik, C. (2025). Deep Learning Methods for Gamma/Hadron Separation in SWGO. In Proceedings of Science. Geneva, CH: Sissa Medialab Srl.

MLA:

Schneider, Markus, et al. "Deep Learning Methods for Gamma/Hadron Separation in SWGO." Proceedings of the 39th International Cosmic Ray Conference, ICRC 2025, Geneva Sissa Medialab Srl, 2025.

BibTeX: Download