SPS Seminar - Discovering radio transients with machine learning and citizen science

When:  Thursday 23 November at 14.00

Where:  Robert Hooke/Teams Online

Speaker:  Alex Anderson (Oxford

Hosted by: Hugh Dickinson

Abstract:

Current and upcoming interferometers can now sample wide swathes of the radio sky with unprecedented sensitivity and cadence. As a result, we can now discover novel radio transients across an immense range of astrophysical regimes - from flare stars to FRBs. I will discuss recent, serendipitous discoveries being made with the MeerKAT radio telescope and how we can make the most of new facilities coming online. This includes how citizen scientists have scoured through commensal data and uncovered 100s of new variable and transient sources. This is the first crowdsourcing project dedicated to radio transients in this manner and has uncovered variable sources as different as nearby flare stars, pulsars and AGN. I will detail unsupervised machine learning techniques being applied to speed up the search for interesting and anomalous sources. These anomaly detection models can, with the use of active learning strategies, be customised to find not only anomalies, but those that are verified as interesting systems for a particular science case. The tested models show great success in recovering transients in our large dataset and these techniques hold great promise for searching large datasets for rare and unique systems.