Hierarchical transfer learning with applications for electricity load forecasting.

Anestis Antoniadis (1) , Solenne Gaucher (2, 3) , Yannig Goude (4)

1 DAO - Données, Apprentissage et Optimisation
2 LMO - Laboratoire de Mathématiques d'Orsay
3 CELESTE - Statistique mathématique et apprentissage
4 EDF R&D - EDF R&D

The recent abundance of data on electricity consumption at different scales opens new challenges and highlights the need for new techniques to leverage information present at finer scales in order to improve forecasts at wider scales. In this work, we take advantage of the similarity between this hierarchical prediction problem and multi-scale transfer learning. We develop two methods for hierarchical transfer learning, based respectively on the stacking of generalized additive models and random forests, and on the use of aggregation of experts. We apply these methods to two problems of electricity load forecasting at national scale, using smart meter data in the first case, and regional data in the second case. For these two usecases, we compare the performances of our methods to that of benchmark algorithms, and we investigate their behaviour using variable importance analysis. Our results demonstrate the interest of both methods, which lead to a significant improvement of the predictions.

Mots clés: Combining forecasts; Time series; Transfer learning; Random forest; Semi-parametric additive model; Demand forecasting

Domaines: Méthodologie [stat.ME]; Machine Learning [stat.ML]; Applications [stat.AP]

Anestis Antoniadis, Solenne Gaucher, Yannig Goude. Hierarchical transfer learning with applications for electricity load forecasting.. 2022. ⟨hal-03429702v3⟩