Background rejection in NEXT using deep neural networks

Abstract

We investigate the potential of using deep learning techniques to reject background events in searches for neutrinoless double beta decay with high pressure xenon time projection chambers capable of detailed track reconstruction. The differences in the topological signatures of background and signal events can be learned by deep neural networks via training over many thousands of events. These networks can then be used to classify further events as signal or background, providing an additional background rejection factor at an acceptable loss of efficiency. The networks trained in this study performed better than previous methods developed based on the use of the same topological signatures by a factor of 1.2 to 1.6, and there is potential for further improvement.

21 pages, 9 figures; formatting changes

References in corpus (18)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• A Unified Approach to the Classical Statistical Analysis of Small Signals
• Going Deeper with Convolutions
• Searching for Exotic Particles in High-Energy Physics with Deep Learning
• Double Beta Decay, Majorana Neutrinos, and Neutrino Mass
• Search for Majorana Neutrinos near the Inverted Mass Hierarchy Region with KamLAND-Zen
• Jet-Images -- Deep Learning Edition
• A Convolutional Neural Network Neutrino Event Classifier
• The event generator DECAY4 for simulation of double beta processes and decay of radioactive nuclei
• Sensitivity of NEXT-100 to neutrinoless double beta decay
• Initial results of NEXT-DEMO, a large-scale prototype of the NEXT-100 experiment
• Operation and first results of the NEXT-DEMO prototype using a silicon photomultiplier tracking array
• Characterisation of NEXT-DEMO using xenon K${}_{\alpha }$ X-rays
• Phenomenology of neutrinoless double beta decay
• Revealing Fundamental Physics from the Daya Bay Neutrino Experiment using Deep Neural Networks
• An homeopathic cure to pure Xenon large diffusion
• The NEXT experiment for neutrinoless double beta decay searches
• Discovery potential of xenon-based neutrinoless double beta decay experiments in light of small angular scale CMB observations

Cited by in corpus (37)

• Neutrinoless Double-Beta Decay: Status and Prospects
• Machine Learning in Nuclear Physics
• Convolutional Neural Networks Applied to Neutrino Events in a Liquid Argon Time Projection Chamber
• A Deep Neural Network for Pixel-Level Electromagnetic Particle Identification in the MicroBooNE Liquid Argon Time Projection Chamber
• Gaseous and dual-phase time projection chambers for imaging rare processes
• Deep Neural Networks for Energy and Position Reconstruction in EXO-200
• Radiopurity assessment of the tracking readout for the NEXT double beta decay experiment
• A Convolutional Neural Network for Multiple Particle Identification in the MicroBooNE Liquid Argon Time Projection Chamber
• Improved Energy Reconstruction in NOvA with Regression Convolutional Neural Networks
• Electromagnetic Showers Beyond Shower Shapes
• Radiogenic backgrounds in the NEXT double beta decay experiment
• Helium-Xenon mixtures to improve topological signature in high pressure gas Xenon TPCs
• Microscopic simulation of xenon-based optical TPCs in the presence of molecular additives
• Context-Enriched Identification of Particles with a Convolutional Network for Neutrino Events
• Radiopurity assessment of the energy readout for the NEXT double beta decay experiment
• Demonstration of background rejection using deep convolutional neural networks in the NEXT experiment
• A Review on Machine Learning for Neutrino Experiments
• Demonstration of the event identification capabilities of the NEXT-White detector
• Topological background discrimination in the PandaX-III neutrinoless double beta decay experiment
• Neutrinoless Double Beta Decay with ${}^{82}$SeF${}_6$ and Direct Ion Imaging
• High Voltage Insulation and Gas Absorption of Polymers in High Pressure Argon and Xenon Gases
• Three-dimensional convolutional neural networks for neutrinoless double-beta decay signal/background discrimination in high-pressure gaseous Time Projection Chamber
• Status and prospects of the NEXT experiment for neutrinoless double beta decay searches
• Timing and characterization of shaped pulses with MHz ADCs in a detector system: a comparative study and deep learning approach
• Pulse shape discrimination using a convolutional neural network for organic liquid scintillator signals
• Time and band-resolved scintillation in time projection chambers based on gaseous xenon
• Double Beta Decay Experiments at Canfranc Underground Laboratory
• Secondary scintillation yield of Xenon with sub-percent levels of CO2 additive: efficiently reducing electron diffusion in HPXe optical TPCs for rare-event detection
• Design and performance of a high-pressure xenon gas TPC as a prototype for a large-scale neutrinoless double-beta decay search
• Interpretable Boosted Decision Tree Analysis for the Majorana Demonstrator
• Report from the A.I. For Nuclear Physics Workshop
• Interpretable machine learning approach for electron antineutrino selection in a large liquid scintillator detector
• Predicting the Masses of Exotic Hadrons with Data Augmentation Using Multilayer Perceptron
• Deep learning based track reconstruction on CEPC luminometer
• Reduction of detection limit and quantification uncertainty due to interferent by neural classification with abstention
• Deep Learning the Effects of Photon Sensors on the Event Reconstruction Performance in an Antineutrino Detector
• Application of CNN to a fine segmented scintillator detector for a single particle and neutrino-nucleon event