Communication Dans Un Congrès Année : 2025

AutoML algorithms for online generalized additive model selection: application to electricity demand forecasting

Apprentissage Automatique pour la sélection de modèles additifs généralisés en ligne : application à la prévision de consommation électrique

Résumé

Electricity demand forecasting is key to ensuring that supply meets demand lest the grid would blackout. Reliable short-term forecasts may be obtained by combining a Generalized Additive Models (GAM) with a State-Space model (Obst et al., 2021), leading to an adaptive (or online) model. A GAM is an over-parameterized linear model defined by a formula and a state-space model involves hyperparameters. Both the formula and adaptation parameters have to be fixed before model training and have a huge impact on the model's predictive performance. We propose optimizing them using the DRAGON package of Keisler (2025), originally designed for neural architecture search. This work generalizes it for automated online generalized additive model selection by defining an efficient modeling of the search space (namely, the space of the GAM formulae and adaptation parameters). Its application to short-term French electricity demand forecasting demonstrates the relevance of the approach.

Fichier principal
Vignette du fichier
Style_and_Template_for_Preprints__arXiv__bio_arXiv_.pdf (489.09 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05020138 , version 1 (04-04-2025)

Licence

Identifiants

  • HAL Id : hal-05020138 , version 1

Citer

Keshav Das, Julie Keisler, Margaux Brégère, Amaury Durand. AutoML algorithms for online generalized additive model selection: application to electricity demand forecasting. AutoML 2025: International Conference on Automated Machine Learning, Sep 2025, New York (NY), United States. pp.23/1--19. ⟨hal-05020138⟩
105 Consultations
246 Téléchargements

Partager

  • More