Domain-informed neural networks for interaction localization within astroparticle experiments

Abstract

This work proposes a domain-informed neural network architecture for experimental particle physics, using particle interaction localization with the time-projection chamber (TPC) technology for dark matter research as an example application. A key feature of the signals generated within the TPC is that they allow localization of particle interactions through a process called reconstruction. While multilayer perceptrons (MLPs) have emerged as a leading contender for reconstruction in TPCs, such a black-box approach does not reflect prior knowledge of the underlying scientific processes. This paper looks anew at neural network-based interaction localization and encodes prior detector knowledge, in terms of both signal characteristics and detector geometry, into the feature encoding and the output layers of a multilayer neural network. The resulting Domain-informed Neural Network (DiNN) limits the receptive fields of the neurons in the initial feature encoding layers in order to account for the spatially localized nature of the signals produced within the TPC. This aspect of the DiNN, which has similarities with the emerging area of graph neural networks in that the neurons in the initial layers only connect to a handful of neurons in their succeeding layer, significantly reduces the number of parameters in the network in comparison to an MLP. In addition, in order to account for the detector geometry, the output layers of the network are modified using two geometric transformations to ensure the DiNN produces localizations within the interior of the detector. The end result is a neural network architecture that has 60% fewer parameters than an MLP, but that still achieves similar localization performance and provides a path to future architectural developments with improved performance because of their ability to encode additional domain knowledge into the architecture.

arXiv comment18 pages, 10 figures. Updated to the published version

References in corpus (34)

• Adam: A Method for Stochastic Optimization
• A Comprehensive Survey on Graph Neural Networks
• Geometric deep learning: going beyond Euclidean data
• The frontier of simulation-based inference
• Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
• Dark Matter Results From 54-Ton-Day Exposure of PandaX-II Experiment
• The STAR Time Projection Chamber: A Unique Tool for Studying High Multiplicity Events at RHIC
• The ALICE TPC, a large 3-dimensional tracking device with fast readout for ultra-high multiplicity events
• Direct Detection of WIMP Dark Matter: Concepts and Status
• Enforcing Analytic Constraints in Neural-Networks Emulating Physical Systems
• Design and Construction of the MicroBooNE Detector
• The Large Underground Xenon (LUX) Experiment
• Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis
• The XENON1T Dark Matter Experiment
• Energy Flow Networks: Deep Sets for Particle Jets
• A Convolutional Neural Network Neutrino Event Classifier
• Sensitivity and discovery potential of the proposed nEXO experiment to neutrinoless double beta decay
• An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution
• First observation of two-neutrino double electron capture in ${}^{124}$Xe with XENON1T
• Convolutional Neural Networks Applied to Neutrino Events in a Liquid Argon Time Projection Chamber
• Observation and applications of single-electron charge signals in the XENON100 experiment
• XENON1T Dark Matter Data Analysis: Signal Reconstruction, Calibration and Event Selection
• Position Reconstruction in a Dual Phase Xenon Scintillation Detector
• A Convolutional Neural Network based Cascade Reconstruction for the IceCube Neutrino Observatory
• Measurement of single electron emission in two-phase xenon
• Semantic Segmentation with a Sparse Convolutional Neural Network for Event Reconstruction in MicroBooNE
• Deep Neural Networks for Energy and Position Reconstruction in EXO-200
• Improved Energy Reconstruction in NOvA with Regression Convolutional Neural Networks
• Analytical Methods for Squaring the Disc
• Detection prospects for the second-order weak decays of ${}^{124}$Xe in multi-tonne xenon time projection chambers
• Machine Learning Accelerated Likelihood-Free Event Reconstruction in Dark Matter Direct Detection
• Horizontal Position Reconstruction in PandaX-II
• HexaConv
• Domain-informed neural networks for interaction localization within astroparticle experiments

Cited by in corpus (3)

• First Dark Matter Search with Nuclear Recoils from the XENONnT Experiment
• Search for New Physics in Electronic Recoil Data from XENONnT
• Domain-informed neural networks for interaction localization within astroparticle experiments