A Convolutional Neural Network based Cascade Reconstruction for the IceCube Neutrino Observatory

Abstract

Continued improvements on existing reconstruction methods are vital to the success of high-energy physics experiments, such as the IceCube Neutrino Observatory. In IceCube, further challenges arise as the detector is situated at the geographic South Pole where computational resources are limited. However, to perform real-time analyses and to issue alerts to telescopes around the world, powerful and fast reconstruction methods are desired. Deep neural networks can be extremely powerful, and their usage is computationally inexpensive once the networks are trained. These characteristics make a deep learning-based approach an excellent candidate for the application in IceCube. A reconstruction method based on convolutional architectures and hexagonally shaped kernels is presented. The presented method is robust towards systematic uncertainties in the simulation and has been tested on experimental data. In comparison to standard reconstruction methods in IceCube, it can improve upon the reconstruction accuracy, while reducing the time necessary to run the reconstruction by two to three orders of magnitude.

arXiv comment39 pages, 15 figures, submitted to Journal of Instrumentation; added references

References in corpus (21)

• Geometric deep learning: going beyond Euclidean data
• Evidence for High-Energy Extraterrestrial Neutrinos at the IceCube Detector
• Neutrino emission from the direction of the blazar TXS 0506+056 prior to the IceCube-170922A alert
• The IceCube Neutrino Observatory: Instrumentation and Online Systems
• Time-integrated Neutrino Source Searches with 10 years of IceCube Data
• Deep Learning and its Application to LHC Physics
• The Design and Performance of IceCube DeepCore
• Calibration and Characterization of the IceCube Photomultiplier Tube
• Energy Reconstruction Methods in the IceCube Neutrino Telescope
• A Convolutional Neural Network Neutrino Event Classifier
• Measurement of South Pole ice transparency with the IceCube LED calibration system
• The IceCube Realtime Alert System
• Convolutional Neural Networks Applied to Neutrino Events in a Liquid Argon Time Projection Chamber
• Application of Deep Learning methods to analysis of Imaging Atmospheric Cherenkov Telescopes data
• Search for Sources of Astrophysical Neutrinos Using Seven Years of IceCube Cascade Events
• Very High-Energy Gamma-Ray Follow-Up Program Using Neutrino Triggers from IceCube
• Efficient propagation of systematic uncertainties from calibration to analysis with the SnowStorm method in IceCube
• Search for astrophysical sources of neutrinos using cascade events in IceCube
• Event reconstruction for KM3NeT/ORCA using convolutional neural networks
• The analysis of VERITAS muon images using convolutional neural networks
• A Review on Machine Learning for Neutrino Experiments

Cited by in corpus (33)

• Observation of high-energy neutrinos from the Galactic plane
• Ultra-High-Energy Cosmic Rays: The Intersection of the Cosmic and Energy Frontiers
• IceCat-1: the IceCube Event Catalog of Alert Tracks
• Graph Neural Networks for Low-Energy Event Classification & Reconstruction in IceCube
• 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
• A deep learning method for the trajectory reconstruction of cosmic rays with the DAMPE mission
• Low Energy Event Reconstruction in IceCube DeepCore
• Deep-learning-based reconstruction of the neutrino direction and energy for in-ice radio detectors
• A flexible event reconstruction based on machine learning and likelihood principles
• A search for an eV-scale sterile neutrino using improved high-energy ${\nu }_{\mu }$ event reconstruction in IceCube
• Optimizing Observables with Machine Learning for Better Unfolding
• Association of the IceCube neutrinos with blazars in the CGRaBS sample
• 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
• Deep learning-based radiointerferometric imaging with GAN-aided training
• Learning Uncertainties the Frequentist Way: Calibration and Correlation in High Energy Physics
• Detector signal characterization with a Bayesian network in XENONnT
• Application of Graph Networks to background rejection in Imaging Air Cherenkov Telescopes
• Separation of track- and shower-like energy deposits in ProtoDUNE-SP using a convolutional neural network
• Bias and Priors in Machine Learning Calibrations for High Energy Physics
• Measuring Oscillations with A Million Atmospheric Neutrinos
• Study of Mass Composition of Cosmic Rays with IceTop and IceCube
• Improved modeling of in-ice particle showers for IceCube event reconstruction
• First Search for Unstable Sterile Neutrinos with the IceCube Neutrino Observatory
• Methods and stability tests associated with the sterile neutrino search using improved high-energy ${\nu }_{\mu }$ event reconstruction in IceCube
• Beyond first light: Global monitoring for high-energy neutrino astronomy
• Accelerating Machine Learning Inference with GPUs in ProtoDUNE Data Processing
• Domain-informed neural networks for interaction localization within astroparticle experiments
• All-Sky Search for Transient Astrophysical Neutrino Emission with 10 Years of IceCube Cascade Events
• Multi-messenger connection in high-energy neutrino astronomy
• Using Convolutional Neural Networks to Reconstruct Energy of GeV Scale IceCube Neutrinos
• Application of Graph Networks to a wide-field Water-Cherenkov-based Gamma-Ray Observatory
• Hybrid-graph neural network method for muon fast reconstruction in neutrino telescopes
• Refine Neutrino Events Reconstruction with BEiT-3