Searching Anomalies Using Nonlinear Dimensionality Reduction Techniques

We have searched for anomalous events using 2,520 hours of archival observations from Murriyang, CSIRO's Parkes radio telescope. These observations were originally undertaken to search for pulsars. We used a machine learning algorithm based on ResNet and Uniform Manifold Approximation and Projection (UMAP) in order to identify parts of the data stream that potentially contain anomalous signals. Many of these anomalous events are radio frequency interference, which were subsequently filtered using multibeam information. We detected 202 anomalous events and provide their positions and event times. One of the events could potentially come from a white dwarf. Some of them may originate from stellar flares, lightning, or be true anomalies, which are currently of an unknown type.