ExoGAN: Retrieving Exoplanetary Atmospheres Using Deep Convolutional Generative Adversarial Networks

Tiziano Zingales, Ingo Peter Waldmann
(Submitted on 7 Jun 2018)

Atmospheric retrievals on exoplanets usually involve computationally intensive Bayesian sampling methods. Large parameter spaces and increasingly complex atmospheric models create a computational bottleneck forcing a trade-off between statistical sampling accuracy and model complexity. This is especially true for upcoming JWST and ARIEL observations. We introduce ExoGAN, the Exoplanet Generative Adversarial Network, a new deep learning algorithm able to recognise molecular features, atmospheric trace-gas abundances and planetary parameters using unsupervised learning. Once trained, ExoGAN is widely applicable to a large number of instruments and planetary types. The ExoGAN retrievals constitute a significant speed improvement over traditional retrievals and can be used either as a final atmospheric analysis or provide prior constraints to a subsequent retrievals.

Comments:    16 pages, 16 figures, 7 tables
Subjects:    Instrumentation and Methods for Astrophysics (astro-ph.IM); Earth and Planetary Astrophysics (astro-ph.EP)
Cite as:    arXiv:1806.02906 [astro-ph.IM] (or arXiv:1806.02906v1 [astro-ph.IM] for this version)
Submission history
From: Tiziano Zingales 
[v1] Thu, 7 Jun 2018 21:19:02 GMT (1045kb,D)
https://arxiv.org/abs/1806.02906