Year: 2024
Type: conference-paper
Abstract: MicroBooNE is a neutrino experiment using a Liquid Argon Time Projection Chamber (LArTPC) detector designed primarily to investigate the anomalous excess of events observed in MiniBooNE.A possible source of the anomaly could be events involving an electron-positron pair, which are difficult to distinguish from electrons in MiniBooNE.Electronpositron pairs from photon pair production are expected to be highly collimated, whereas non-collimated pairs could be evidence for exciting new physics beyond the standard model.We aim to reliably reconstruct the opening angle for electron-positron events using simulations that accurately model the operating conditions of MicroBooNE.To this end, we use machine learning methods, specifically the PointNet++ graph-neural network architecture.We find that the results show great promise, and we describe future work to further investigate potential avenues of improvement, especially on more realistic and complex datasets.
DOI: 10.2172/2426632
Authors: Sowjanya Gollapinni, David Yang
Institution: Los Alamos National Laboratory
Language: English
Topic: Neutrino Physics Research
Subfield: Nuclear and High Energy Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4401436459