A Convolutional Neural Network for Multiple Particle Identification in the MicroBooNE Liquid Argon Time Projection Chamber

TL;DR

This paper introduces MPID, a CNN that classifies multiple particles in MicroBooNE's LArTPC data, improving particle identification for neutrino analysis.

Contribution

The paper presents a novel CNN architecture for multiple particle classification in LArTPC data, extending previous single particle identification methods.

Findings

High accuracy in particle classification on simulated data

Effective performance demonstrated on real detector data

Enhances neutrino event analysis in MicroBooNE

Abstract

We present the multiple particle identification (MPID) network, a convolutional neural network (CNN) for multiple object classification, developed by MicroBooNE. MPID provides the probabilities of $e^-$, $\gamma$, $\mu^-$, $\pi^\pm$, and protons in a single liquid argon time projection chamber (LArTPC) readout plane. The network extends the single particle identification network previously developed by MicroBooNE. MPID takes as input an image either cropped around a reconstructed interaction vertex or containing only activity connected to a reconstructed vertex, therefore relieving the tool from inefficiencies in vertex finding and particle clustering. The network serves as an important component in MicroBooNE's deep learning based $\nu_e$ search analysis. In this paper, we present the network's design, training, and performance on simulation and data from the MicroBooNE detector.