Supervised Machine Learning Methods with Uncertainty Quantification for Exoplanet Atmospheric Retrievals from Transmission Spectroscopy
This work addresses the need for faster atmospheric retrievals in exoplanet research, particularly for upcoming observatories like JWST, but is incremental as it systematically compares existing methods rather than introducing new ones.
The paper tackled the computational expense of standard Bayesian retrievals for exoplanet atmospheric parameters from transmission spectroscopy by benchmarking several existing machine learning regression techniques, finding that the best-performing combination achieved efficient and robust performance validated on JWST observations of WASP-39b.
Standard Bayesian retrievals for exoplanet atmospheric parameters from transmission spectroscopy, while well understood and widely used, are generally computationally expensive. In the era of the JWST and other upcoming observatories, machine learning approaches have emerged as viable alternatives that are both efficient and robust. In this paper we present a systematic study of several existing machine learning regression techniques and compare their performance for retrieving exoplanet atmospheric parameters from transmission spectra. We benchmark the performance of the different algorithms on the accuracy, precision, and speed. The regression methods tested here include partial least squares (PLS), support vector machines (SVM), k nearest neighbors (KNN), decision trees (DT), random forests (RF), voting (VOTE), stacking (STACK), and extreme gradient boosting (XGB). We also investigate the impact of different preprocessing methods of the training data on the model performance. We quantify the model uncertainties across the entire dynamical range of planetary parameters. The best performing combination of ML model and preprocessing scheme is validated on a the case study of JWST observation of WASP-39b.