Estimating Galactic Distances From Images Using Self-supervised Representation Learning
arXiv:2101.04293
Abstract
We use a contrastive self-supervised learning framework to estimate distances to galaxies from their photometric images. We incorporate data augmentations from computer vision as well as an application-specific augmentation accounting for galactic dust. We find that the resulting visual representations of galaxy images are semantically useful and allow for fast similarity searches, and can be successfully fine-tuned for the task of redshift estimation. We show that (1) pretraining on a large corpus of unlabeled data followed by fine-tuning on some labels can attain the accuracy of a fully-supervised model which requires 2-4x more labeled data, and (2) that by fine-tuning our self-supervised representations using all available data labels in the Main Galaxy Sample of the Sloan Digital Sky Survey (SDSS), we outperform the state-of-the-art supervised learning method.
References in corpus (11)
- Measuring Reddening with SDSS Stellar Spectra and Recalibrating SFD
- A Simple Framework for Contrastive Learning of Visual Representations
- The Eleventh and Twelfth Data Releases of the Sloan Digital Sky Survey: Final Data from SDSS-III
- Improved Baselines with Momentum Contrastive Learning
- Big Self-Supervised Models are Strong Semi-Supervised Learners
- Slicing Through Multicolor Space: Galaxy Redshifts From Broadband Photometry
- Self-training with Noisy Student improves ImageNet classification
- Photometric redshifts from SDSS images using a Convolutional Neural Network
- Photometric redshifts for the SDSS Data Release 12
- An automatic taxonomy of galaxy morphology using unsupervised machine learning
- Fixing the train-test resolution discrepancy: FixEfficientNet