Year: 2024
Type: conference-paper
Abstract: This presentation describes the process of training NuGraph2, a Graphical Neural Network for event reconstruction, on simulated ICARUS neutrino event data. This began with an investigation into filtering ICARUS spacepoint data. Then NuGraph2 was repeatedly trained on three event samples, which were used for finding optimized machine-learning parameters and to find and fix the causes of several crashes in NuGraph2 s preprocessing and training scripts.
DOI: 10.2172/2482066
Author: Eric Novello
Institution: Fermi National Accelerator Laboratory
Language: English
Topic: Neutrino Physics Research
Subfield: Nuclear and High Energy Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4405582156