ExoGAN: Retrieving Exoplanetary Atmospheres Using Deep Convolutional Generative Adversarial Networks

Abstract

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. It 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 subsequent retrieval.

arXiv comment19 pages, 17 figures, 7 tables

References in corpus (27)

• The James Webb Space Telescope
• Multimodal nested sampling: an efficient and robust alternative to MCMC methods for astronomical data analysis
• Catalog of Nearby Exoplanets
• A Temperature and Abundance Retrieval Method for Exoplanet Atmospheres
• Improved parameters for extrasolar transiting planets
• A Systematic Retrieval Analysis of Secondary Eclipse Spectra I: A Comparison of Atmospheric Retrieval Techniques
• Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
• Helium in the eroding atmosphere of an exoplanet
• A population study of gaseous exoplanets
• How to Distinguish between Cloudy Mini-Neptunes and Water/Volatile-Dominated Super-Earths
• Tau-REx I: A next generation retrieval code for exoplanetary atmospheres
• A chemical model for the atmosphere of hot Jupiters
• Generative Adversarial Networks recover features in astrophysical images of galaxies beyond the deconvolution limit
• A consistent retrieval analysis of 10 Hot Jupiters observed in transmission
• Tau-REx II: Retrieval of emission spectra
• HELIOS-Retrieval: An Open-source, Nested Sampling Atmospheric Retrieval Code, Application to the HR 8799 Exoplanets and Inferred Constraints for Planet Formation
• Scientific Discovery with the James Webb Space Telescope
• Searching for Exoplanets Using Artificial Intelligence
• Exploring Biases of Atmospheric Retrievals in Simulated JWST Transmission Spectra of Hot Jupiters
• Retrieval of Exoplanet Emission Spectra with HyDRA
• A HST/WFC3 Thermal Emission Spectrum of the Hot Jupiter HAT-P-7b
• Use of the MultiNest algorithm for gravitational wave data analysis
• Evidence for a Dayside Thermal Inversion and High Metallicity for the Hot Jupiter WASP-18b
• Dreaming of atmospheres
• A comparative study of WASP-67b and HAT-P-38b from WFC3 data
• PSFGAN: a generative adversarial network system for separating quasar point sources and host galaxy light
• Transit Clairvoyance: Enhancing TESS follow-up using artificial neural networks

Cited by in corpus (34)

• Forward modelling and retrievals with PLATON, a fast open source tool
• Atmospheric Regimes and Trends on Exoplanets and Brown Dwarfs
• 2022 Review of Data-Driven Plasma Science
• Aurora: A Generalised Retrieval Framework for Exoplanetary Transmission Spectra
• An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval
• Learning to Denoise Astronomical Images with U-nets
• Outstanding Challenges of Exoplanet Atmospheric Retrievals
• Neural posterior estimation for exoplanetary atmospheric retrieval
• Interpreting High-Resolution Spectroscopy of Exoplanets Using Cross-Correlations and Supervised Machine Learning
• Machine learning inference of the interior structure of low-mass exoplanets
• ExoMDN: Rapid characterization of exoplanet interior structures with Mixture Density Networks
• Accurate Machine Learning Atmospheric Retrieval via a Neural Network Surrogate Model for Radiative Transfer
• Assessment of Supervised Machine Learning for Atmospheric Retrieval of Exoplanets
• Painting halos from cosmic density fields of dark matter with physically motivated neural networks
• Aura-3D: A Three-dimensional Atmospheric Retrieval Framework for Exoplanet Transmission Spectra
• Pulsar Candidate Identification Using Semi-Supervised Generative Adversarial Networks
• A Catalogue of Exoplanet Atmospheric Retrieval Codes
• Convolutional neural networks as an alternative to Bayesian retrievals
• Detrending Exoplanetary Transit Light Curves with Long Short-Term Memory Networks
• PyLightcurve-torch: a transit modelling package for deep learning applications in PyTorch
• Hierarchical Bayesian Atmospheric Retrieval Modeling for Population Studies of Exoplanet Atmospheres: A Case Study on the Habitable Zone
• Estimating dayside effective temperatures of hot Jupiters and associated uncertainties through Gaussian process regression
• Terminus: A Versatile Simulator for Space-based Telescopes
• How do we optimally sample model grids of exoplanet spectra?
• Grid-Based Atmospheric Retrievals for Reflected-Light Spectra of Exoplanets using PSGnest
• Multilayer Perceptron and Geometric Albedo Spectra for Quick Parameter Estimations of Exoplanets
• On the Dayside Atmosphere of WASP-12b
• Flow Matching for Atmospheric Retrieval of Exoplanets: Where Reliability meets Adaptive Noise Levels
• Near-instantaneous Atmospheric Retrievals and Model Comparison with FASTER
• Radiative Transfer and Inversion codes for characterizing planetary atmospheres: an overview
• A Machine Learning Approach to Enhancing eROSITA Observations
• Transverse Vector Decomposition Method for Analytical Inversion of Exoplanet Transit Spectra
• TelescopeML -- I. An End-to-End Python Package for Interpreting Telescope Datasets through Training Machine Learning Models, Generating Statistical Reports, and Visualizing Results
• $\text{Exoformer}$: Accelerating Bayesian atmospheric retrievals with transformer neural networks