Pion Identification through Machine Learning for the MOLLER Experiment at the Thomas Jefferson National Accelerator Facility
Date Thesis Awarded
Honors Thesis -- Access Restricted On-Campus Only
Bachelors of Science (BS)
Dr. David Armstrong, Dr. Phiala Shanahan
This report describes the implementation of a deep neural network for particle identification on the MOLLER experiment. The MOLLER experiment, currently in its early stages of design at the Thomas Jefferson National Accelerator Facility (JLab), will attempt to measure the parity-violating asymmetry present in the elastic electron-electron scattering, to a precision of 0.7 ppb. While the Standard Model precisely predicts this asymmetry, if the value measured by the MOLLER experiment were to differ significantly from the predicted value, then the experiment could provide laboratory-based evidence of physics beyhond the Standard Model (BSM) and point researchers in the right direction for its exploration. The high energy electron beam used in this experiment is predicted to generate scattered electrons as well as a background of roughly 0.13 percent pions. While the ratio of pions to electrons will be small, their presence may significantly affect the asymmetry measurement. The detected particles, predominantly pions and electrons, must thus be classified. Here, an algorithm is proposed to classify particles detected in the MOLLER experiment using deep neural networks (DNNs). Once the classification algorithm is successfully written and proven to work, the uncertainty in the classification of the particles as pions, electrons, or positrons will be determined. If successful, this classification algorithm may be used to optimize the design of the experiment hardware.
Burns, Anne-Katherine, "Pion Identification through Machine Learning for the MOLLER Experiment at the Thomas Jefferson National Accelerator Facility" (2019). Undergraduate Honors Theses. William & Mary. Paper 1310.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
On-Campus Access Only