A Likelihood-Based Generative Approach for Spatially Consistent Precipitation Downscaling
Not on the map. Why?
Abstract (arXiv)
Deep learning has emerged as a promising tool for precipitation downscaling. However, current models rely on likelihood-based loss functions to properly model the precipitation distribution, leading to spatially inconsistent projections when sampling. This work explores a novel approach by fusing the strengths of likelihood-based and adversarial losses used in generative models. As a result, we propose a likelihood-based generative approach for precipitation downscaling, leveraging the benefits of both methods.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 11 of its 45 references that are in the map. First posted on arXiv 26 Jun 2024.
Builds on
11 of 45 references.
References · 11
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
- 2017Improved Training of Wasserstein GANscited by 2,235
- 2014Conditional Generative Adversarial Netscited by 2,162
- 2019Deep Learning for Image Super-resolution: A Surveycited by 369
- 2021Skillful Precipitation Nowcasting using Deep Generative Models of Radarcited by 311
- 2017DeepSD: Generating High Resolution Climate Change Projections through Single Image Super-Resolutioncited by 82
- 2020Stochastic Super-Resolution for Downscaling Time-Evolving Atmospheric Fields with a Generative Adversarial Networkcited by 70
- 2022A Generative Deep Learning Approach to Stochastic Downscaling of Precipitation Forecastscited by 66
- 2022Increasing the accuracy and resolution of precipitation forecasts using deep generative modelscited by 35
- 2020TRU-NET: A Deep Learning Approach to High Resolution Prediction of Rainfallcited by 19
- 2023On the use of Deep Generative Models for Perfect Prognosis Climate Downscalingcited by 2
Built on by
Citing papers per year: 2024: 1.
1 paper.