Semi-supervised Learning of Galaxy Morphology using Equivariant Transformer Variational Autoencoders
This work aims to automate galaxy morphology classification for astronomers and astrophysicists, reducing the need for extensive human labelling efforts.
This paper addresses the challenge of classifying galaxy morphology given the rapid increase in unlabelled galaxy images. The authors developed a Variational Autoencoder (VAE) with Equivariant Transformer layers and a classifier network, demonstrating improved accuracy on the Galaxy Zoo dataset and achieving higher accuracy with fewer labels compared to existing approaches through pre-training.
The growth in the number of galaxy images is much faster than the speed at which these galaxies can be labelled by humans. However, by leveraging the information present in the ever growing set of unlabelled images, semi-supervised learning could be an effective way of reducing the required labelling and increasing classification accuracy. We develop a Variational Autoencoder (VAE) with Equivariant Transformer layers with a classifier network from the latent space. We show that this novel architecture leads to improvements in accuracy when used for the galaxy morphology classification task on the Galaxy Zoo data set. In addition we show that pre-training the classifier network as part of the VAE using the unlabelled data leads to higher accuracy with fewer labels compared to exiting approaches. This novel VAE has the potential to automate galaxy morphology classification with reduced human labelling efforts.