Neutrino characterisation using convolutional neural networks in CHIPS water Cherenkov detectors

Abstract This work presents a novel approach to water Cherenkov neutrino detector event reconstruction and classification. Three forms of Convolutional Neural Network have been trained reject cosmic muon events, classify beam estimate energies, using only slightly modified version the raw as input. When evaluated on realistic selection simulated CHIPS-5kton prototype this new significantly increases performance over standard likelihood-based simple neural network