Machine Learning Applications for Waveform Analysis

Since the later 20th century, the search for physics beyond the Standard Model (BSM) has been paramount to many nuclear and particle physicists. Neutron and nuclear beta decay experiments provide one avenue to search for evidence of BSM physics by contributing to the unitarity check of the Cabibbo-Kobayashi-Maskawa matrix. Many of these experiments detect neutron decay products as digitized waveforms. As computing power increases and novel algorithms are developed, it is compelling to investigate machine learning methods as an analytic tool for such waveform data. These methods can allow for very fast data exploration techniques, and if pseudodata is available predictive models can be built for tasks such as particle identification. This thesis will report machine learning analysis done for both the Ca-45 Beta Spectrum Measurement at LANL and the BL2 Neutron Lifetime Measurement at NIST.