First Deep Learning based Event Reconstruction for Low-Energy Excess Searches with MicroBooNE
This paper describes algorithms developed to isolate and accurately reconstruct two-track νμ-like events that are contained within the MicroBooNE detector. This reconstruction has applications to searches for neutrino oscillations and measurements of cross sections using events that are chargedcurrent quasi-elastic-like, among other applications. The algorithms we discuss will be applicable to all detectors running in Fermilab’s SBN program, and any future LArTPC experiment with beam energies ∼ 1 GeV.
Paper
Full text
First Deep Learning based Event Reconstruction for Low-Energy Excess Searches with MicroBooNE
Semantic Scholar · Physics · 2018
Abstract
This paper describes algorithms developed to isolate and accurately reconstruct two-track νμ-like events that are contained within the MicroBooNE detector. This reconstruction has applications to searches for neutrino oscillations and measurements of cross sections using events that are chargedcurrent quasi-elastic-like, among other applications. The algorithms we discuss will be applicable to all detectors running in Fermilab’s SBN program, and any future LArTPC experiment with beam energies ∼ 1 GeV.