Deep Non-Crossing Quantiles through the Partial Derivative

Axel Brando, Joan Gimeno, José A. Rodríguez-Serrano, Jordi Vitrià

Producció científica: Article en revista indexadaArticle de conferènciaAvaluat per experts

5 Cites (Scopus)

Resum

Quantile Regression (QR) provides a way to approximate a single conditional quantile. To have a more informative description of the conditional distribution, QR can be merged with deep learning techniques to simultaneously estimate multiple quantiles. However, the minimisation of the QR-loss function does not guarantee non-crossing quantiles, which affects the validity of such predictions and introduces a critical issue in certain scenarios. In this article, we propose a generic deep learning algorithm for predicting an arbitrary number of quantiles that ensures the quantile monotonicity constraint up to the machine precision and maintains its modelling performance with respect to alternative models. The presented method is evaluated over several real-world datasets obtaining state-of-the-art results as well as showing that it scales to large-size data sets.

Idioma originalAnglès
Pàgines (de-a)7902-7914
Nombre de pàgines13
RevistaProceedings of Machine Learning Research
Volum151
Estat de la publicacióPublicada - 2022
Publicat externament
Esdeveniment25th International Conference on Artificial Intelligence and Statistics, AISTATS 2022 - Virtual, Online, Spain
Durada: 28 de març 202230 de març 2022

Fingerprint

Navegar pels temes de recerca de 'Deep Non-Crossing Quantiles through the Partial Derivative'. Junts formen un fingerprint únic.

Com citar-ho