Astronomy-Themed GAN Papers and Codes

GANs or generative adversarial networks (Goodfellow et al. 2014), often dubbed as the "coolest idea in machine learning in the last 20 years" has sparked a major interest in the use of generative models for science applications. Here is a list of science-related (with a focus on astronomy/cosmology) GAN (and a few variational autoencoder) resources. The list will be updated as new papers/codes come out.

Example of GAN generated art (source)

 

1. cosmoGAN: (git) (paper)
2. radioGAN: (paper)
3. exoGAN: (git) (paper)
4. Enabling Dark Energy Science with Deep Generative Models of Galaxy Images: (paper)
5. Fast Cosmic Web Simulations with Generative Adversarial Networks: (paper)
6. Generative Adversarial Networks Recover Features in Astrophysical Images of Galaxies Beyond the Deconvolution Limit: (paper)
7. Forging New Worlds: High-Resolution Synthetic Galaxies With Chained Generative Adversarial Networks: (git) (paper)
8. Stellar Cluster Detection using GMM with Deep Variational Autoencoder: (paper)
9. A Deep Generative Model for Astronomical Images of Galaxies: (paper)
10. Generative Deep Fields: Arbitrarily Sized, Random Synthetic Astronomical Images Through Deep Learning: (git)
11. Model Comparison of Dark Energy models Using Deep Networks (paper)
12. Self-supervised Learning with Physics-aware Neural Networks I: Galaxy Model Fitting (paper)
13. CMB-GAN: Fast Simulations of Cosmic Microwave Background Anisotropy Maps Using Deep Learning (paper)
14. Cosmological N-body Simulations: a Challenge for Scalable Generative Models (paper) (git)