Details

Autor(en) / Beteiligte

• Akbar, F
• Ghosh, Anushree
• Young, Steven
• Akhter, S
• Dar, Z Ahmad
• Ansari, V
• Ascencio, M. V.
• Athar, M Sajjad
• Bodek, A
• Bonilla, JL
• Bravar, A
• Budd, H
• Caceres, G
• Cai, T
• Carneiro, MF
• Diaz, GA
• Felix, J
• Fields, L
• Filkins, A
• Fine, R
• Gaur, PK
• Gran, R
• Harris, DA
• Jena, Debdeep
• Jena, S
• Kleykamp, J
• Klustova, A
• Last, D
• Lozano, A
• Lu, X
• Maher, E
• Manly, S
• Mann, WA
• Farland, KS
• Messerly, B
• Miller, Jonathan
• Moreno, O
• Morfin, JG
• Nelson, JK
• Nguyen, C
• Olivier, A
• Paolone, V
• Perdue, Gabriel
• Plows, K
• Ramirez, MA
• Ruterbories, D
• Su, H
• Syrotenko, VS
• Waldron, AV
• Yaeggy, B
• Zazueta, L

Titel

Vertex finding in neutrino-nucleus interaction: a model architecture comparison

Ist Teil von

• Journal of instrumentation, 2022-08, Vol.17 (8)

Ort / Verlag

United States: Institute of Physics (IOP)

Erscheinungsjahr

2022

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Here we compare different neural network architectures for machine learning algorithms designed to identify the neutrino interaction vertex position in the MINERvA detector. The architectures developed and optimized by hand are compared with the architectures developed in an automated way using the package "Multi-node Evolutionary Neural Networks for Deep Learning" (MENNDL), developed at Oak Ridge National Laboratory. While the domain-expert hand-tuned network was the best performer, the differences were negligible and the auto-generated networks performed as well. There is always a trade-off between human, and computer resources for network optimization and this work suggests that automated optimization, assuming resources are available, provides a compelling way to save significant expert time.