Optimizing exoplanet atmosphere retrieval using unsupervised machine-learning classification

Abstract

One of the principal bottlenecks to atmosphere characterisation in the era of all-sky surveys is the availability of fast, autonomous and robust atmospheric retrieval methods. We present a new approach using unsupervised machine learning to generate informed priors for retrieval of exoplanetary atmosphere parameters from transmission spectra. We use principal component analysis (PCA) to efficiently compress the information content of a library of transmission spectra forward models generated using the PLATON package. We then apply a $k$-means clustering algorithm in PCA space to segregate the library into discrete classes. We show that our classifier is almost always able to instantaneously place a previously unseen spectrum into the correct class, for low-to-moderate spectral resolutions, $R$, in the range $R=30-300$ and noise levels up to $10$~per~cent of the peak-to-trough spectrum amplitude. The distribution of physical parameters for all members of the class therefore provides an informed prior for standard retrieval methods such as nested sampling. We benchmark our informed-prior approach against a standard uniform-prior nested sampler, finding that our approach is up to a factor two faster, with negligible reduction in accuracy. We demonstrate the application of this method to existing and near-future observatories, and show that it is suitable for real-world application. Our general approach is not specific to transmission spectroscopy and should be more widely applicable to cases that involve repetitive fitting of trusted high-dimensional models to large data catalogues, including beyond exoplanetary science.

arXiv commentAccepted for publication in MNRAS

References in corpus (27)

• Scikit-learn: Machine Learning in Python
• Detection of an Extrasolar Planet Atmosphere
• Astrophysics Source Code Library
• Clouds in the atmosphere of the super-Earth exoplanet GJ 1214b
• The orbital motion, absolute mass, and high-altitude winds of exoplanet HD209458b
• petitRADTRANS: a Python radiative transfer package for exoplanet characterization and retrieval
• Sodium Absorption From the Exoplanetary Atmosphere of HD189733b Detected in the Optical Transmission Spectrum
• The signature of orbital motion from the dayside of the planet tau Bootis b
• Detection of Carbon Monoxide and Water Absorption Lines in an Exoplanet Atmosphere
• Detection of water absorption in the day side atmosphere of HD 189733 b using ground-based high-resolution spectroscopy at 3.2 microns
• Tau-REx I: A next generation retrieval code for exoplanetary atmospheres
• Atomic iron and titanium in the atmosphere of the exoplanet KELT-9b
• Spectrally resolved helium absorption from the extended atmosphere of a warm Neptune-mass exoplanet
• Detection of titanium oxide in the atmosphere of a hot Jupiter
• Tau-REx II: Retrieval of emission spectra
• High-Resolution Spectroscopic Detection of TiO and Stratosphere in the Day-side of WASP-33b
• Gran Telescopio Canarias OSIRIS Transiting Exoplanet Atmospheric Survey: Detection of potassium in XO-2b from narrowband spectrophotometry
• A library of ATMO forward model transmission spectra for hot Jupiter exoplanets
• Forward modelling and retrievals with PLATON, a fast open source tool
• Exo-Transmit: An Open-Source Code for Calculating Transmission Spectra for Exoplanet Atmospheres of Varied Composition
• VLT FORS2 comparative transmission spectroscopy: Detection of Na in the atmosphere of WASP-39b from the ground
• Exoplanet atmospheres with GIANO II. Detection of molecular absorption in the dayside spectrum of HD 102195b
• Evidence for multiple molecular species in the hot Jupiter HD 209458b
• Exoplanet Spectroscopy and Photometry with the Twinkle Space Telescope
• An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval
• GTC OSIRIS transiting exoplanet atmospheric survey: detection of potassium in HAT-P-1b from narrowband spectrophotometry
• Exoplanetary atmosphere target selection in the era of comparative planetology

Cited by in corpus (17)

• PLATON II: New Capabilities And A Comprehensive Retrieval on HD 189733b Transit and Eclipse Data
• Atmospheric Regimes and Trends on Exoplanets and Brown Dwarfs
• Aurora: A Generalised Retrieval Framework for Exoplanetary Transmission Spectra
• Neural posterior estimation for exoplanetary atmospheric retrieval
• Alkaline Exospheres of Exoplanet Systems: Evaporative Transmission Spectra
• A review of unsupervised learning in astronomy
• Accurate Machine Learning Atmospheric Retrieval via a Neural Network Surrogate Model for Radiative Transfer
• Assessment of Supervised Machine Learning for Atmospheric Retrieval of Exoplanets
• Data mining techniques on astronomical spectra data. I : Clustering Analysis
• Hierarchical Bayesian Atmospheric Retrieval Modeling for Population Studies of Exoplanet Atmospheres: A Case Study on the Habitable Zone
• Terminus: A Versatile Simulator for Space-based Telescopes
• Follow the Water: Finding Water, Snow and Clouds on Terrestrial Exoplanets with Photometry and Machine Learning
• Flow Matching for Atmospheric Retrieval of Exoplanets: Where Reliability meets Adaptive Noise Levels
• Inferring Exoplanet Disequilibria with Multivariate Information in Atmospheric Reaction Networks
• Radiative Transfer and Inversion codes for characterizing planetary atmospheres: an overview
• Machine learning based stellar classification with highly sparse photometry data
• $\text{Exoformer}$: Accelerating Bayesian atmospheric retrievals with transformer neural networks