english Icono del idioma   español Icono del idioma  

Please use this identifier to cite or link to this item: https://hdl.handle.net/20.500.12008/48652 How to cite
Title: A CRPS loss for deep probabilistic regression.
Authors: Marchesoni-Acland, Franco
Alonso-Suárez, Rodrigo
Herrera, Andrés
Kherroubi, Josselin
Morel, Jean M.
Facciolo, Gabriele
Type: Preprint
Keywords: CRPS, Probabilistic Regression, Deep Learning, Solar Forecasting
Issue Date: 2024
Abstract: Probabilistic regression is relevant in highstakes areas such as energy forecasting, financial risk assessment, or healthcare. Deep models that directly output a probability distribution usually use ensembles or frame regression as classification into bins. In contrast, we propose to optimize directly for the Continuous Ranked Probability Score (CRPS), a proper scoring rule for probabilistic predictions. For the flexible histogram-like distributions, the CRPS is differentiable and can be used as the loss function of any deep model. We derive and implement the CRPS loss and showcase its performance against cross-entropy in a solar forecasting application. This new loss enables anyone to easily make any deep regressor probabilistic by simply using the new loss with the same computational cost. Surprisingly, using the CRPS loss provides superior results even when training a deterministic regressor. Code and data available at github.com/franchesoni/differentiable-crps.
Citation: Marchesoni-Acland, F., Alonso-Suárez, R., Herrera, A. y otros. A CRPS loss for deep probabilistic regression. [Preprint] Publicado en: Proceedings del IEEE URUCON, 18-20 Nov. 2024. 4 p. DOI: 10.1109/URUCON63440.2024.10850406.
License: Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0)
Appears in Collections:Publicaciones académicas y científicas - Laboratorio de Energía Solar (LES)

Files in This Item:
File Description SizeFormat  
MAHKMF24.pdfPreprint2,62 MBAdobe PDFView/Open


This item is licensed under a Creative Commons License Creative Commons