To Sample or Not To Sample: Retrieving Exoplanetary Spectra with Variational Inference and Normalising Flows

TL;DR

This paper introduces a novel neural network framework combining normalising flows and a differentiable forward model for efficient, high-fidelity atmospheric retrieval of exoplanets, significantly reducing computation time and enabling Bayesian model selection.

Contribution

The authors develop a neural network-based atmospheric retrieval method that requires only a single observation, offers high-fidelity posteriors, reduces computation by 75%, and performs Bayesian model selection.

Findings

Neural network requires only one observation for training.

Produces posterior distributions comparable to sampling-based methods.

Reduces forward model computation by 75%.

Abstract

Current endeavours in exoplanet characterisation rely on atmospheric retrieval to quantify crucial physical properties of remote exoplanets from observations. However, the scalability and efficiency of the technique are under strain with increasing spectroscopic resolution and forward model complexity. The situation becomes more acute with the recent launch of the James Webb Space Telescope and other upcoming missions. Recent advances in Machine Learning provide optimisation-based Variational Inference as an alternative approach to perform approximate Bayesian Posterior Inference. In this investigation we combined Normalising Flow-based neural network with our newly developed differentiable forward model, Diff-Tau, to perform Bayesian Inference in the context of atmospheric retrieval. Using examples from real and simulated spectroscopic data, we demonstrated the superiority of our proposed framework: 1) Training Our neural network only requires a single observation; 2) It produces high-fidelity posterior distributions similar to sampling-based retrieval and; 3) It requires 75% less forward model computation to converge. 4.) We performed, for the first time, Bayesian model selection on our trained neural network. Our proposed framework contribute towards the latest development of a neural-powered atmospheric retrieval. Its flexibility and speed hold the potential to complement sampling-based approaches in large and complex data sets in the future.