Radio Galaxy Zoo: compact and extended radio source classification with deep learning

Abstract

Machine learning techniques have been increasingly useful in astronomical applications overthe last few years, for example in the morphological classification of galaxies. Convolutionalneural networks have proven to be highly effective in classifying objects in image data. Inthe context of radio-interferometric imaging in astronomy, we looked for ways to identifymultiple components of individual sources. To this effect, we design a convolutional neuralnetwork to differentiate between different morphology classes using sources from the RadioGalaxy Zoo (RGZ) citizen science project. In this first step, we focus on exploring the factorsthat affect the performance of such neural networks, such as the amount of training data, number and nature of layers, and the hyperparameters. We begin with a simple experiment inwhich we only differentiate between two extreme morphologies, using compact and multiplecomponentextended sources. We found that a three-convolutional layer architecture yieldedvery good results, achieving a classification accuracy of 97.4 per cent on a test data set. The same architecture was then tested on a four-class problem where we let the networkclassify sources into compact and three classes of extended sources, achieving a test accuracyof 93.5 per cent. The best-performing convolutional neural network set-up has been verifiedagainst RGZ Data Release 1 where a final test accuracy of 94.8 per cent was obtained, usingboth original and augmented images. The use of sigma clipping does not offer a significantbenefit overall, except in cases with a small number of training images.