Quantile-Regression-Ensemble: A Deep Learning Algorithm for Downscaling Extreme Precipitation

4Citations
Citations of this article
14Readers
Mendeley users who have this article in their library.

Abstract

Global Climate Models (GCMs) simulate low resolution climate projections on a global scale. The native resolution of GCMs is generally too low for societal-level decisionmaking. To enhance the spatial resolution, downscaling is often applied to GCM output. Statistical downscaling techniques, in particular, are well-established as a cost-effective approach. They require significantly less computational time than physics-based dynamical downscaling. In recent years, deep learning has gained prominence in statistical downscaling, demonstrating significantly lower error rates compared to traditional statistical methods. However, a drawback of regression-based deep learning techniques is their tendency to overfit to the mean sample intensity. Extreme values as a result are often underestimated. Problematically, extreme events have the largest societal impact. We propose Quantile- Regression-Ensemble (QRE), an innovative deep learning algorithm inspired by boosting methods. Its primary objective is to avoid trade-offs between fitting to sample means and extreme values by training independent models on a partitioned dataset. Our QRE is robust to redundant models and not susceptible to explosive ensemble weights, ensuring a reliable training process. QRE achieves lower Mean Squared Error (MSE) compared to various baseline models. In particular, our algorithm has a lower error for high-intensity precipitation events over New Zealand, highlighting the ability to represent extreme events accurately.

Cite

CITATION STYLE

APA

Bailie, T., Koh, Y. S., Rampal, N., & Gibson, P. B. (2024). Quantile-Regression-Ensemble: A Deep Learning Algorithm for Downscaling Extreme Precipitation. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 38, pp. 21914–21922). Association for the Advancement of Artificial Intelligence. https://doi.org/10.1609/aaai.v38i20.30193

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free