Convolutional neural networks as an alternative to Bayesian retrievals

Abstract

Exoplanet observations are currently analysed with Bayesian retrieval techniques. Due to the computational load of the models used, a compromise is needed between model complexity and computing time. Analysis of data from future facilities, will need more complex models which will increase the computational load of retrievals, prompting the search for a faster approach for interpreting exoplanet observations. Our goal is to compare machine learning retrievals of exoplanet transmission spectra with nested sampling, and understand if machine learning can be as reliable as Bayesian retrievals for a statistically significant sample of spectra while being orders of magnitude faster. We generate grids of synthetic transmission spectra and their corresponding planetary and atmospheric parameters, one using free chemistry models, and the other using equilibrium chemistry models. Each grid is subsequently rebinned to simulate both HST/WFC3 and JWST/NIRSpec observations, yielding four datasets in total. Convolutional neural networks (CNNs) are trained with each of the datasets. We perform retrievals on a 1,000 simulated observations for each combination of model type and instrument with nested sampling and machine learning. We also use both methods to perform retrievals on real WFC3 transmission spectra. Finally, we test how robust machine learning and nested sampling are against incorrect assumptions in our models. CNNs reach a lower coefficient of determination between predicted and true values of the parameters. Nested sampling underestimates the uncertainty in ~8% of retrievals, whereas CNNs estimate them correctly. For real WFC3 observations, nested sampling and machine learning agree within $2\sigma$ for ~86% of spectra. When doing retrievals with incorrect assumptions, nested sampling underestimates the uncertainty in ~12% to ~41% of cases, whereas this is always below ~10% for the CNN.

Accepted for publication in A&A

References in corpus (10)

• The James Webb Space Telescope
• Hot Exoplanet Atmospheres Resolved with Transit Spectroscopy (HEARTS) IV. A spectral inventory of atoms and molecules in the high-resolution transmission spectrum of WASP-121 b
• The Transit Light Source Effect II: The Impact of Stellar Heterogeneity on Transmission Spectra of Planets Orbiting Broadly Sun-like Stars
• The ARCiS framework for Exoplanet Atmospheres: Modelling Philosophy and Retrieval
• Sparkling nights and very hot days on WASP-18b: the formation of clouds and the emergence of an ionosphere
• An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval
• Forecasting the Impact of Stellar Activity on Transiting Exoplanet Spectra
• Machine learning inference of the interior structure of low-mass exoplanets
• Assessment of Supervised Machine Learning for Atmospheric Retrieval of Exoplanets
• Aluminium oxide in the atmosphere of hot Jupiter WASP-43b

Cited by in corpus (12)

• Neural posterior estimation for exoplanetary atmospheric retrieval
• The ESO SupJup Survey I: Chemical and isotopic characterisation of the late L-dwarf DENIS J0255-4700 with CRIRES${}^{+}$
• Machine learning-accelerated chemistry modeling of protoplanetary disks
• Grid-Based Atmospheric Retrievals for Reflected-Light Spectra of Exoplanets using PSGnest
• Flow Matching for Atmospheric Retrieval of Exoplanets: Where Reliability meets Adaptive Noise Levels
• Parameterizing pressure-temperature profiles of exoplanet atmospheres with neural networks
• The ESO SupJup Survey VIII. Chemical fingerprints of young L dwarf twins
• Near-instantaneous Atmospheric Retrievals and Model Comparison with FASTER
• Radiative Transfer and Inversion codes for characterizing planetary atmospheres: an overview
• Reconstructing Atmospheric Parameters of Exoplanets Using Deep Learning
• Transverse Vector Decomposition Method for Analytical Inversion of Exoplanet Transit Spectra
• Deciphering transmission spectra by exploring the solar paradigm