Solar Axion Search in SNO+ with Machine Learning
This event is part of the Preliminary Oral Exam.
Examining Committee:
Chris Grant(PI, High Energy Experiment) Indara Suarez (High Energy Experiment), Claudio Rebbi(High Energy Theory), Mark Kon(BU Math Dept.)
Abstract:
Liquid scintillator-based (LS) detectors are one of the leading detector technologies in the search for neutrinoless double-beta decay. It has been demonstrated that such a detector can also be adopted to search for Axion Like Particle from solar. They are currently limited by naturally occurring and spallation induced backgrounds. Here we use a convolutional neural network, a common algorithm from computer vision, to attempt to distinguish between events that would have made it through existing cuts. We train our network on Monte Carlo simulated truth data with a range of detector capabilities, and evaluate the training results in these different conditions. The ultimate goal of this project is to apply sophisticated machine learning techniques to reject backgrounds in real detector data.