The Atacama Cosmology Telescope: the Development of Machine Learning Tools for Detecting Millimeter Sources in Timestream Pre-processing

We present a new machine learning algorithm for classifying short-duration features in raw time ordered data (TODs) of cosmic microwave background survey observations. The algorithm, specifically designed for the Atacama Cosmology Telescope (ACT), works in conjunction with the previous TOD preprocessing techniques that employ statistical thresholding to indiscriminately remove all large spikes in the data, whether they are due to noise features, cosmic rays, or true astrophysical sources in a process called "data cuts". This has the undesirable effect of excising real astrophysical sources, including transients, from the data. The machine learning algorithm demonstrated in this work uses the output from these data cuts and is able to differentiate between electronic noise, cosmic rays, and point sources, enabling the removal of undesired signals while retaining true astrophysical signals during TOD pre-processing. We achieve an overall accuracy of 90 spikes of different origin and, importantly, 94 by astrophysical sources. Our algorithm also measures the amplitude of any detected source seen more than once and produces a sub-minute to minute light curve, providing information on its short timescale variability. This automated algorithm for source detection and amplitude estimation will be particularly useful for upcoming surveys with large data volumes, such as the Simons Observatory.