Speaker
Description
The upgraded near neutrino detector of the T2K experiment is briefly described. The central part is the 3D scintillator fine-grained detector (SuperFGD). The detector comprised of about 2 million 1 × 1 × 1 cm3 plastic scintillator cubes with three orthogonal holes for wavelength shifting fiber readout. It is used as an active neutrino target to detect charged particles and photons. A combination of SuperFGD with the other subdetector systems in the ND280 detector provides a possibility to detect more neutrino events (due to an additional active target), as well as to increase the accuracy of the neutrino events reconstruction.
In charged-current muon-neutrino analysis, the outgoing muon is selected to reconstruct the neutrino energy. In SuperFGD, however, pions can be misidentified as muons, complicating this selection. This work uses a Graph Neural Network (GNN) to identify muons and pions. We present the efficiency and purity of the GNN-based classification of these particles and discuss its impact on event selections. The results are based on Monte Carlo simulations of neutrino interactions in the upgraded ND280 detector.
-
K. Abe et al. (T2K Collaboration), The T2K experiment, Nucl. Instrum. Meth. A 659 (2011) 106-135, DOI: 10.1016/j.nima.2011.06.067
-
Yu. Kudenko et al. (T2K Collaboration), The near neutrino detector of the T2K experiment, Nucl. Instrum. Meth. A 598 (2009) 289-295, DOI: 10.1016/j.nima.2008.08.029
-
Abe S. et al., The super fine-grained detector for the T2K neutrino oscillation experiment. Nucl.Instrum.Meth.A 1092 (2026) 171882
-
Abe K. [et al.]. T2K ND280 Upgrade – Technical Design Report, JINST 6. 2011. P12004 [arXiv:1901.03750 ] Jan. 11 2019, 196 p.
-
Shlomi, J., Battaglia, P., & Vlimant, J.-R. "Graph neural networks in particle physics." Machine Learning: Science and Technology 2.2 (2021): 021001.
