| Metadato | Etiqueta | Valor | Idioma |
|---|---|---|---|
| dc.contributor.author | Autor | Marchesoni-Acland, Franco | - |
| dc.contributor.author | Autor | Alonso-Suárez, Rodrigo | - |
| dc.contributor.author | Autor | Herrera, Andrés | - |
| dc.contributor.author | Autor | Kherroubi, Josselin | - |
| dc.contributor.author | Autor | Morel, Jean M. | - |
| dc.contributor.author | Autor | Facciolo, Gabriele | - |
| dc.contributor.filiacion | Filiación | Marchesoni-Acland Franco | - |
| dc.contributor.filiacion | Filiación | Alonso-Suárez Rodrigo | - |
| dc.contributor.filiacion | Filiación | Herrera Andrés | - |
| dc.contributor.filiacion | Filiación | Kherroubi Josselin | - |
| dc.contributor.filiacion | Filiación | Morel Jean M. | - |
| dc.contributor.filiacion | Filiación | Facciolo Gabriele | - |
| dc.date.accessioned | Fecha ingreso | 2025-03-12T17:14:49Z | - |
| dc.date.available | Fecha disponible | 2025-03-12T17:14:49Z | - |
| dc.date.issued | Año de publicación | 2024 | - |
| dc.description.abstract | Resumen | 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. | es |
| dc.format.extent | Extensión | 4 p. | es |
| dc.format.mimetype | Formato | application/pdf | es |
| dc.identifier.citation | Citación | 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. | es |
| dc.identifier.uri | URI | https://hdl.handle.net/20.500.12008/48652 | - |
| dc.language.iso | Idioma | en | es |
| dc.rights | Derechos | Las obras depositadas en el Repositorio se rigen por la Ordenanza de los Derechos de la Propiedad Intelectual de la Universidad de la República.(Res. Nº 91 de C.D.C. de 8/III/1994 – D.O. 7/IV/1994) y por la Ordenanza del Repositorio Abierto de la Universidad de la República (Res. Nº 16 de C.D.C. de 07/10/2014) | es |
| dc.rights.licence | Licencia | Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0) | es |
| dc.subject | Palabras clave | CRPS | es |
| dc.subject | Palabras clave | Probabilistic Regression | es |
| dc.subject | Palabras clave | Deep Learning | es |
| dc.subject | Palabras clave | Solar Forecasting | es |
| dc.title | Título | A CRPS loss for deep probabilistic regression. | es |
| dc.type | Tipo de documento | Preprint | es |
Colecciones:
| Ficheros | Descripción | Tamaño | Formato | ||
|---|---|---|---|---|---|
| MAHKMF24.pdf | Preprint | 2.56 MB | Adobe PDF |