Scalable Deep Convolutional Neural Networks for Sparse, Locally Dense Liquid Argon Time Projection Chamber Data

Abstract

Deep convolutional neural networks (CNNs) show strong promise for analyzing scientific data in many domains including particle imaging detectors such as a liquid argon time projection chamber (LArTPC). Yet the high sparsity of LArTPC data challenges traditional CNNs which were designed for dense data such as photographs. A naive application of CNNs on LArTPC data results in inefficient computations and a poor scalability to large LArTPC detectors such as the Short Baseline Neutrino Program and Deep Underground Neutrino Experiment. Recently Submanifold Sparse Convolutional Networks (SSCNs) have been proposed to address this challenge. We report their performance on a 3D semantic segmentation task on simulated LArTPC samples. In comparison with standard CNNs, we observe that the computation memory and wall-time cost for inference are reduced by factor of 364 and 33 respectively without loss of accuracy. The same factors for 2D samples are found to be 93 and 3.1 respectively. Using SSCN, we present the first machine learning-based approach to the reconstruction of Michel electrons using public 3D LArTPC samples. We find a Michel electron identification efficiency of 93.9% with 96.7% of true positive rate. Reconstructed Michel electron clusters yield 95.4% in average pixel clustering efficiency and 95.5% in purity. The results are compelling to show strong promise of scalable data reconstruction technique using deep neural networks for large scale LArTPC detectors.

Corrected URL to dataset Corrected figures and numbers

References in corpus (1)

• A Deep Neural Network for Pixel-Level Electromagnetic Particle Identification in the MicroBooNE Liquid Argon Time Projection Chamber

Cited by in corpus (19)

• Semantic Segmentation with a Sparse Convolutional Neural Network for Event Reconstruction in MicroBooNE
• GPU-accelerated machine learning inference as a service for computing in neutrino experiments
• Demonstration of background rejection using deep convolutional neural networks in the NEXT experiment
• A Review on Machine Learning for Neutrino Experiments
• Graph Neural Network for Object Reconstruction in Liquid Argon Time Projection Chambers
• Application of Transfer Learning to Neutrino Interaction Classification
• Pulse shape discrimination using a convolutional neural network for organic liquid scintillator signals
• Point Proposal Network for Reconstructing 3D Particle Endpoints with Sub-Pixel Precision in Liquid Argon Time Projection Chambers
• Scalable, End-to-End, Deep-Learning-Based Data Reconstruction Chain for Particle Imaging Detectors
• Accelerating Deep Neural Networks for Real-time Data Selection for High-resolution Imaging Particle Detectors
• Cosmic Background Removal with Deep Neural Networks in SBND
• PILArNet: Public Dataset for Particle Imaging Liquid Argon Detectors in High Energy Physics
• Deep Learning strategies for ProtoDUNE raw data denoising
• Neutrino Characterisation using Convolutional Neural Networks in CHIPS water Cherenkov detectors
• Adversarial methods to reduce simulation bias in neutrino interaction event filtering at Liquid Argon Time Projection Chambers
• Suppression of accidental backgrounds with deep neural networks in the PandaX-II experiment
• Contrastive Learning for Robust Representations of Neutrino Data
• Closing the Neutrino "BSM Gap": Physics Potential of Atmospheric Through-Going Muons at DUNE
• LArTPC hit-based topology classification with quantum machine learning and symmetry