Semi-supervised Learning of Galaxy Morphology using Equivariant Transformer Variational Autoencoders

TL;DR

This paper introduces a semi-supervised Variational Autoencoder with Equivariant Transformer layers that improves galaxy morphology classification accuracy and reduces labeling efforts by leveraging unlabelled data.

Contribution

It presents a novel VAE architecture with equivariant transformers and demonstrates its effectiveness in semi-supervised galaxy classification tasks.

Findings

Improved accuracy on Galaxy Zoo dataset.

Pre-training with unlabelled data enhances performance with fewer labels.

Potential to automate galaxy morphology classification.

Abstract

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.