Self-Supervised Representation Learning for Astronomical Images

From WebRef.org
Revision as of 12:52, 25 December 2020 by Webref (talk | contribs) (Created page with "Md Abul Hayat, George Stein, Peter Harrington, Zarija Lukić, Mustafa Mustafa Sky surveys are the largest data generators in astronomy, making automated tools for extracting...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

Md Abul Hayat, George Stein, Peter Harrington, Zarija Lukić, Mustafa Mustafa

Sky surveys are the largest data generators in astronomy, making automated tools for extracting meaningful scientific information an absolute necessity. We show that, without the need for labels, self-supervised learning recovers representations of sky survey images that are semantically useful for a variety of scientific tasks. These representations can be directly used as features, or fine-tuned, to outperform supervised methods trained only on labeled data. We apply a contrastive learning framework on multi-band galaxy photometry from the Sloan Digital Sky Survey (SDSS) to learn image representations. We then use them for galaxy morphology classification, and fine-tune them for photometric redshift estimation, using labels from the Galaxy Zoo 2 dataset and SDSS spectroscopy. In both downstream tasks, using the same learned representations, we outperform the supervised state-of-the-art results, and we show that our approach can achieve the accuracy of supervised models while using 2-4 times fewer labels for training.

https://arxiv.org/abs/2012.13083


Sponsor: Shop Our Screen Protectors Designed Specifically For Your Laptop - Only At EyeJust