Neutrinos are the least understood fundamental particles. They provide a unique window to the inner workings of our universe, and they are one of our best hopes for seeing new physics beyond the standard model.
Sterile neutrinos are a hypothetical fourth kind of neutrino that have been proposed to explain anomalies in short base-line neutrino oscillation experiments. In order to make sense of these anomalies, along with the results of other experiments that do not see sterile neutrinos, I performed global fits to data from many different experiments. We have published fits to 3+1 and 3+2 models, as well as a fit that includes IceCube data and constrains the 3+1 mixing matrix for the first time.
During my thesis research, I was a member of the IceCube south pole neutrino telescope collaboration. I studied the uncertainty in the neutrino flux for the sterile neutrino search. Along with Nick Rodd and Ben Safdi of MIT, I looked for evidence of point sources of neutrinos. The discovery of astronomical objects that emit high-energy neutrinos would open an entirely new view of our universe. I am also applied deep learning to identify tau neutrinos in IceCube data, and assisted with a search for Lorentz Violation.
While a University Research Association Fermilab fellow, I helped with the testing and construction of the MicroBooNE experiment at Fermilab. I also investigated how deep learning can be used to classify neutrino events in MicroBooNE.
With the assistance of the American-Australian association, I am applied cutting edge research in deep learning to the problem of neutrino classification and reconstruction. Deep learning has the potential to meet the challenge of processing the enormous amount of data that are produced by current and future generation neutrino detectors.
