Near-instantaneous Atmospheric Retrievals and Model Comparison with FASTER

TL;DR

FASTER is a neural-network based method that enables near-instantaneous atmospheric retrievals and model comparisons for exoplanet spectra, significantly reducing computational costs compared to traditional Bayesian techniques.

Contribution

Introduces FASTER, a neural network framework for rapid atmospheric retrievals and Bayesian model comparison, scalable to large spectral datasets from telescopes like JWST.

Findings

FASTER matches nested sampling in parameter estimation accuracy.

Enables real-time analysis of thousands of spectra.

Provides insights into model dependencies and uncertainties.

Abstract

In the era of the James Webb Space Telescope (JWST), the dramatic improvement in the spectra of exoplanetary atmospheres demands a corresponding leap forward in our ability to analyze them: atmospheric retrievals need to be performed on thousands of spectra, applying to each large ensembles of models (that explore atmospheric chemistry, thermal profiles and cloud models) to identify the best one(s). In this limit, traditional Bayesian inference methods such as nested sampling become prohibitively expensive. We introduce FASTER (Fast Amortized Simulation-based Transiting Exoplanet Retrieval), a neural-network based method for performing atmospheric retrieval and Bayesian model comparison at a fraction of the computational cost of classical techniques. We demonstrate that the marginal posterior distributions of all parameters within a model as well as the posterior probabilities of the models we consider match those computed using nested sampling both on mock spectra, and for the real NIRSpec PRISM spectrum of WASP-39b. The true power of the FASTER framework comes from its amortized nature, which allows the trained networks to perform practically instantaneous Bayesian inference and model comparison over ensembles of spectra -- real or simulated -- at minimal additional computational cost. This offers valuable insight into the expected results of model comparison (e.g., distinguishing cloudy from cloud-free and isothermal from non-isothermal models), as well as their dependence on the underlying parameters, which is computationally unfeasible with nested sampling. This approach will constitute as large a leap in spectral analysis as the original retrieval methods based on Markov Chain Monte Carlo have proven to be.