Big-data challenges associated with large radio surveys, such as the Evolutionary Map of the Universe (EMU) conducted with the ASKAP telescope, will require non-traditional approaches. We have applied self-supervised algorithms for discovering new radio sources such as Odd Radio Circles. 

I will discuss how we use weakly-supervised deep learning techniques for detecting extended radio sources, along with the integration of novel semi/fully supervised object detection methods to identify, categorize and group complex radio galaxies with multiple components. These novel computer vision methods produce value-added catalogues of radio galaxies which are essential for astronomical and cosmological studies.