Application of Graph Networks to a wide-field Water-Cherenkov-based Gamma-Ray Observatory

Abstract

With their wide field of view and high duty cycle, water-Cherenkov-based observatories are integral to studying the very high-energy gamma-ray sky. For gamma-ray observations, precise event reconstruction and highly effective background rejection are crucial and have been continuously improving in recent years. In this work, we investigate the application of graph neural networks (GNNs) to background rejection and energy reconstruction and benchmark their performance against state-of-the-art methods. In our simulation study, we find that GNNs outperform hand-designed classification algorithms and observables in background rejection and find an improved energy resolution compared to template-based methods.

arXiv commentpublished version, 23 pages, 14 figures, 4 tables

References in corpus (34)

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Geometric deep learning: going beyond Euclidean data
• Observations of the Crab Nebula with H.E.S.S
• ParticleNet: Jet Tagging via Particle Clouds
• Observation of the Crab Nebula with the HAWC Gamma-Ray Observatory
• Observation of high-energy neutrinos from the Galactic plane
• The First VERITAS Telescope
• Performance of the MAGIC stereo system obtained with Crab Nebula data
• Implementation of the Random Forest Method for the Imaging Atmospheric Cherenkov Telescope MAGIC
• A Monte Carlo Template based analysis for Air-Cherenkov Arrays
• Observation of TeV Gamma Rays from the Crab Nebula with Milagro Using a New Background Rejection Technique
• Gamma-Hadron Separation in Very-High-Energy gamma-ray astronomy using a multivariate analysis method
• A Deep Learning-based Reconstruction of Cosmic Ray-induced Air Showers
• Milagrito: a TeV air-shower array
• The High-Altitude Water Cherenkov (HAWC) Observatory in México: The Primary Detector
• Improved $\gamma$/hadron separation for the detection of faint gamma-ray sources using boosted decision trees
• A Convolutional Neural Network based Cascade Reconstruction for the IceCube Neutrino Observatory
• Graph Neural Networks for Low-Energy Event Classification & Reconstruction in IceCube
• The Southern Wide-Field Gamma-Ray Observatory (SWGO): A Next-Generation Ground-Based Survey Instrument for VHE Gamma-Ray Astronomy
• Event reconstruction for KM3NeT/ORCA using convolutional neural networks
• Deep-Learning based Reconstruction of the Shower Maximum $X_{\max }$ using the Water-Cherenkov Detectors of the Pierre Auger Observatory
• Extraction of the Muon Signals Recorded with the Surface Detector of the Pierre Auger Observatory Using Recurrent Neural Networks
• Automatic Design of Artificial Neural Networks for Gamma-Ray Detection
• Inference of the Mass Composition of Cosmic Rays with energies from $10^{18.5}$ to $10^{20}$ eV using the Pierre Auger Observatory and Deep Learning
• Muon identification in a compact single-layered water Cherenkov detector and gamma/hadron discrimination using Machine Learning techniques
• Identification of Patterns in Cosmic-Ray Arrival Directions using Dynamic Graph Convolutional Neural Networks
• Treatment of Errors in Efficiency Calculations
• Measurement of the Depth of Maximum of Air-Shower Profiles with energies between $10^{18.5}$ and $10^{20}$ eV using the Surface Detector of the Pierre Auger Observatory and Deep Learning
• Gamma/Hadron Separation with the HAWC Observatory
• A Template-based $\gamma$-ray Reconstruction Method for Air Shower Arrays
• Application of Graph Networks to background rejection in Imaging Air Cherenkov Telescopes
• Performance of LHAASO-WCDA and Observation of Crab Nebula as a Standard Candle
• Neural networks for gamma-hadron separation in MAGIC