Ensembling geophysical models with Bayesian Neural Networks

Sengupta, Ushnish; Amos, Matt; Hosking, Scott ORCID: https://orcid.org/0000-0002-3646-3504; Rasmussen, Carl Edward; Juniper, Matthew P.; Young, Paul J.. 2020 Ensembling geophysical models with Bayesian Neural Networks. Advances in Neural Information Processing Systems, 33. 13, pp.

Before downloading, please read NORA policies.

Text NeurIPS-2020-ensembling-geophysical-models-with-bayesian-neural-networks-Paper.pdf Restricted to NORA staff only Download (1MB) | Request a copy

Abstract/Summary

Ensembles of geophysical models improve projection accuracy and express uncertainties. We develop a novel data-driven ensembling strategy for combining geophysical models using Bayesian Neural Networks, which infers spatiotemporally varying model weights and bias while accounting for heteroscedastic uncertainties in the observations. This produces more accurate and uncertainty-aware projections without sacrificing interpretability. Applied to the prediction of total column ozone from an ensemble of 15 chemistry-climate models, we find that the Bayesian neural network ensemble (BayNNE) outperforms existing ensembling methods, achieving a 49.4% reduction in RMSE for temporal extrapolation, and a 67.4% reduction in RMSE for polar data voids, compared to a weighted mean. Uncertainty is also well-characterized, with 90.6% of the data points in our extrapolation validation dataset lying within 2 standard deviations and 98.5% within 3 standard deviations.

Item Type:	Publication - Article
Date made live:	15 Mar 2021 11:42 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/528739

Actions (login required)

View Item

Document Downloads

More statistics for this item...