Articles | Volume 11, issue 8
Atmos. Meas. Tech., 11, 4627–4643, 2018
https://doi.org/10.5194/amt-11-4627-2018
Atmos. Meas. Tech., 11, 4627–4643, 2018
https://doi.org/10.5194/amt-11-4627-2018

Research article 09 Aug 2018

Research article | 09 Aug 2018

A neural network approach to estimating a posteriori distributions of Bayesian retrieval problems

Simon Pfreundschuh et al.

Model code and software

typhon - Tools for atmospheric research John Mrziglod, Lukas Kluft, Oliver Lemke, Gerrit Holl, Simon Pfreundschuh, Richard Larsson, Takayoshi Yamada, and Jakob Doerr https://doi.org/10.5281/zenodo.1300319

Download
Short summary
A novel neural-network-based retrieval method is proposed that combines the flexibility and computational efficiency of machine learning retrievals with the consistent treatment of uncertainties of Bayesian methods. Numerical experiments are presented that show the consistency of the proposed method with the Bayesian formulation as well as its ability to represent non-Gaussian retrieval errors. With this, the proposed method overcomes important limitations of traditional methods.