Neutrino Characterisation using Convolutional Neural Networks in CHIPS water Cherenkov detectors
arXiv:2206.14904 · doi:10.1088/1748-0221/18/06/P06032
Abstract
This work presents a novel approach to water Cherenkov neutrino detector event reconstruction and classification. Three forms of a Convolutional Neural Network have been trained to reject cosmic muon events, classify beam events, and estimate neutrino energies, using only a slightly modified version of the raw detector event as input. When evaluated on a realistic selection of simulated CHIPS-5kton prototype detector events, this new approach significantly increases performance over the standard likelihood-based reconstruction and simple neural network classification.
45 pages, 22 figures, 5 tables, to be submitted to Nuclear Instruments and Methods in Physics Research - section A
References in corpus (8)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Measurements of neutrino oscillation in appearance and disappearance channels by the T2K experiment with 6.6E20 protons on target
- Sensitivity of Super-Kamiokande with Gadolinium to Low Energy Anti-neutrinos from Pre-supernova Emission
- A Convolutional Neural Network for Multiple Particle Identification in the MicroBooNE Liquid Argon Time Projection Chamber
- Convolutional Neural Networks for Shower Energy Prediction in Liquid Argon Time Projection Chambers
- Deep-Learning-Based Kinematic Reconstruction for DUNE
- Deep Learning strategies for ProtoDUNE raw data denoising
- CHIPS Event Reconstruction and Design Optimisation