Self-Supervised Representation Learning for Astronomical Images

Abstract

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.

arXiv commentThe codes, trained models, and data can be found at https://portal.nersc.gov/project/dasrepo/self-supervised-learning-sdss

References in corpus (40)

• Measuring Reddening with SDSS Stellar Spectra and Recalibrating SFD
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• A Simple Framework for Contrastive Learning of Visual Representations
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• The Eleventh and Twelfth Data Releases of the Sloan Digital Sky Survey: Final Data from SDSS-III
• The 2.5 m Telescope of the Sloan Digital Sky Survey
• Improved Baselines with Momentum Contrastive Learning
• Spectroscopic Target Selection in the Sloan Digital Sky Survey: The Main Galaxy Sample
• The Sloan Digital Sky Survey Photometric Camera
• Galaxy Zoo : Morphologies derived from visual inspection of galaxies from the Sloan Digital Sky Survey
• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• Euclid Definition Study Report
• Drizzle: A Method for the Linear Reconstruction of Undersampled Images
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• Galaxy Zoo 2: detailed morphological classifications for 304,122 galaxies from the Sloan Digital Sky Survey
• What Makes for Good Views for Contrastive Learning?
• Deep Clustering for Unsupervised Learning of Visual Features
• Big Self-Supervised Models are Strong Semi-Supervised Learners
• Using Pre-Training Can Improve Model Robustness and Uncertainty
• Preparing Red-Green-Blue (RGB) Images from CCD Data
• Improving galaxy morphologies for SDSS with Deep Learning
• Slicing Through Multicolor Space: Galaxy Redshifts From Broadband Photometry
• Galaxy Zoo : 'Hanny's Voorwerp', a quasar light echo?
• Photometric redshifts from SDSS images using a Convolutional Neural Network
• Photometric redshifts for the SDSS Data Release 12
• Photometric redshift estimation via deep learning
• Galaxy Zoo: Probabilistic Morphology through Bayesian CNNs and Active Learning
• The Galaxy Zoo survey for giant AGN-ionized clouds: past and present black-hole accretion events
• Galaxy Zoo: comparing the demographics of spiral arm number and a new method for correcting redshift bias
• Transfer learning for galaxy morphology from one survey to another
• An automatic taxonomy of galaxy morphology using unsupervised machine learning
• Galaxy morphological classification in deep-wide surveys via unsupervised machine learning
• Identifying Strong Lenses with Unsupervised Machine Learning using Convolutional Autoencoder
• Beyond the Hubble Sequence -- Exploring Galaxy Morphology with Unsupervised Machine Learning
• Detecting outliers in astronomical images with deep generative networks
• AstroVaDEr: Astronomical Variational Deep Embedder for Unsupervised Morphological Classification of Galaxies and Synthetic Image Generation
• Deep Learning at Scale for the Construction of Galaxy Catalogs in the Dark Energy Survey
• Pushing automated morphological classifications to their limits with the Dark Energy Survey
• A Study of the Point Spread Function in SDSS Images
• Morphology in the Era of Large Surveys

Cited by in corpus (11)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Astronomia ex machina: a history, primer, and outlook on neural networks in astronomy
• Practical Galaxy Morphology Tools from Deep Supervised Representation Learning
• Mining for Strong Gravitational Lenses with Self-supervised Learning
• Optimized Photometric Redshifts for the Cosmic Assembly Near-Infrared Deep Extragalactic Legacy Survey (CANDELS)
• Self-supervised Anomaly Detection for New Physics
• Capturing the physics of MaNGA galaxies with self-supervised Machine Learning
• Machine learning technique for morphological classification of galaxies from SDSS. II. The image-based morphological catalogs of galaxies at 0.02<z<0.1
• Galaxy Spin Classification I: Z-wise vs S-wise Spirals With Chirality Equivariant Residual Network
• An Image Processing approach to identify solar plages observed at 393.37 nm by the Kodaikanal Solar Observatory
• Lenses In VoicE (LIVE): Searching for strong gravitational lenses in the VOICE@VST survey using Convolutional Neural Networks