Self-supervised Learning for Astronomical Image Classification
This addresses the challenge of small labeled datasets in astronomy, offering a domain-specific solution that is incremental over existing pre-training methods.
The paper tackles the problem of limited labeled data in astronomical image classification by proposing a self-supervised pre-training technique using unlabeled images, which outperforms ImageNet pre-training in many cases.
In Astronomy, a huge amount of image data is generated daily by photometric surveys, which scan the sky to collect data from stars, galaxies and other celestial objects. In this paper, we propose a technique to leverage unlabeled astronomical images to pre-train deep convolutional neural networks, in order to learn a domain-specific feature extractor which improves the results of machine learning techniques in setups with small amounts of labeled data available. We show that our technique produces results which are in many cases better than using ImageNet pre-training.