Neutral pion reconstruction using machine learning in the experiment at ⟨Eν⟩ 6 GeV. (29th July 2021)

• Record Type: Journal Article
• Authors: Ghosh, A.; Yaeggy, B.; Galindo, R.; Ahmad Dar, Z.; Akbar, F.; Ascencio, M.V.; Bashyal, A.; Bercellie, A.; Bonilla, J.L.; Caceres, G.; Cai, T.; Carneiro, M.F.; da Motta, H.; Díaz, G.A.; Felix, J.; Filkins, A.; Fine, R.; Gago, A.M.; Golan, T.; Gran, R.; Harris, D.A.; Henry, S.; Jena, S.; Jena, D.; Kleykamp, J.; Kordosky, M.; Last, D.; Le, T.; Lozano, A.; Lu, X.-G.; Maher, E.; Manly, S.; Mann, W.A.; Mauger, C.; McFarland, K.S.; Messerly, B.; Miller, J.; Montano, L.M.; Naples, D.; Nelson, J.K.; Nguyen, C.; Olivier, A.; Paolone, V.; Perdue, G.N.; Ramírez, M.A.; Ray, H.; Ruterbories, D.; Solano Salinas, C.J.; Su, H.; Sultana, M.; Syrotenko, V.S.; Valencia, E.; Wospakrik, M.; Wret, C.; Yang, K.; Zazueta, L.;
• Is Part Of: Journal of instrumentation. Volume 16:Number 7(2021)
• Publication Date: 2021-07-29
• Subjects: Analysis and statistical methods -- Pattern recognition, cluster finding, calibration and fitting methods -- Neutrino detectors; Scientific apparatus and instruments -- Periodicals; 502.84
• DOI: 10.1088/1748-0221/16/07/P07060
• Languages: English
• ISSNs: 1748-0221

Abstract

Abstract: This paper presents a novel neutral-pion reconstruction that takes advantage of the machine learning technique of semantic segmentation using MINERvA data collected between 2013–2017, with an average neutrino energy of 6 GeV. Semantic segmentation improves the purity of neutral pion reconstruction from two γs from 70.7 ± 0.9% to 89.3 ± 0.7% and improves the efficiency of the reconstruction by approximately 40%. We demonstrate our method in a charged current neutral pion production analysis where a single neutral pion is reconstructed. This technique is applicable to modern tracking calorimeters, such as the new generation of liquid-argon time projection chambers, exposed to neutrino beams with ⟨ E ν ⟩ between 1–10 GeV. In such experiments it can facilitate the identification of ionization hits which are associated with electromagnetic showers, thereby enabling improved reconstruction of charged-current νe events arising from νμ → νe appearance.