Pulse Shape Discrimination with Machine learning at JSNS2

The JSNS2 (J-PARC Sterile Neutrino Search at the J-PARC Spallation Neutron Source) experiment searches for neutrino oscillations at 24m baseline from the J-PARC’s 3 GeV 1 MW proton beam incident on a mercury target at the MLF. The particles from the mercury target are detected at the JSNS2 detector which is filled with the gadolinium (Gd) -loaded Liquid Scintillator (LS). The Fast Neutron (FN) is one of the important backgrounds in this experiment. It is important to have a good performance of the Pulse Shape Discrimination (PSD) algorithm to reduce Fast Neutron backgrounds. We developed an improved PSD algorithm based on the Convolutional Neural Network (CNN) for the JSNS2 to discriminate Michel Electrons and Fast Neutrons in the JSNS2 experiment. In this presentation, our CNN architecture and its PSD performance with the real data taken in 2021 will be shown