Application of Machine Learning for Low Energy Electron LArTPC Reconstruction

Detection of Low Energy Neutrino interactions in a liquid argon time-projection chamber (LArTPC) rely are reliant on detection of both ionization electrons and scintillation – emitted photons. The scintillation photons provide the necessary timing information to properly reconstruct the event. Efficiency of photon detection decreases with the energy level of the interaction. Well understood mechanisms for electron diffusion within the LArTPC can be utilized to accurately predict the drift coordinate in place of the scintillation data. A machine learning algorithm is to be trained on simulation data to deduce the appropriate timing values from electron drift, eliminating the need for photodetection, and overall simplifying the detection apparatus.