English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Current endeavours in exoplanet characterisation rely on atmosphericretrieval to quantify crucial physical properties of remote exoplanets fromobservations. However, the scalability and efficiency of the technique areunder strain with increasing spectroscopic resolution and forward modelcomplexity. The situation becomes more acute with the recent launch of theJames Webb Space Telescope and other upcoming missions. Recent advances inMachine Learning provide optimisation-based Variational Inference as analternative approach to perform approximate Bayesian Posterior Inference. Inthis investigation we combined Normalising Flow-based neural network with ournewly developed differentiable forward model, Diff-Tau, to perform BayesianInference in the context of atmospheric retrieval. Using examples from real andsimulated spectroscopic data, we demonstrated the superiority of our proposedframework: 1) Training Our neural network only requires a single observation;2) It produces high-fidelity posterior distributions similar to sampling-basedretrieval and; 3) It requires 75% less forward model computation to converge.4.) We performed, for the first time, Bayesian model selection on our trainedneural network. Our proposed framework contribute towards the latestdevelopment of a neural-powered atmospheric retrieval. Its flexibility andspeed hold the potential to complement sampling-based approaches in large andcomplex data sets in the future.

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