AutoML Algorithms for Online Generalized Additive Model Selection: Application to Electricity Demand Forecasting

Keshav Das, Julie Keisler, Margaux Brégère, Amaury Durand
Proceedings of the Fourth International Conference on Automated Machine Learning, PMLR 293:23/1-19, 2025.

Abstract

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, 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 to optimize them using automated Machine Learning. For this purpose, we define an efficient modeling of the search space (namely, the space of the GAM formulae and adaptation parameters) as well as mutation and crossover operators on this space and apply an evolutionary algorithm to select the best parameters. We evaluate our method on short-term French electricity demand forecasting which demonstrates the relevance of the approach.

Cite this Paper

BibTeX
@InProceedings{pmlr-v293-das25a, title = {AutoML Algorithms for Online Generalized Additive Model Selection: Application to Electricity Demand Forecasting}, author = {Das, Keshav and Keisler, Julie and Br\'eg\`ere, Margaux and Durand, Amaury}, booktitle = {Proceedings of the Fourth International Conference on Automated Machine Learning}, pages = {23/1--19}, year = {2025}, editor = {Akoglu, Leman and Doerr, Carola and van Rijn, Jan N. and Garnett, Roman and Gardner, Jacob R.}, volume = {293}, series = {Proceedings of Machine Learning Research}, month = {08--11 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v293/main/assets/das25a/das25a.pdf}, url = {https://proceedings.mlr.press/v293/das25a.html}, abstract = {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, 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 to optimize them using automated Machine Learning. For this purpose, we define an efficient modeling of the search space (namely, the space of the GAM formulae and adaptation parameters) as well as mutation and crossover operators on this space and apply an evolutionary algorithm to select the best parameters. We evaluate our method on short-term French electricity demand forecasting which demonstrates the relevance of the approach.} }
Endnote
%0 Conference Paper %T AutoML Algorithms for Online Generalized Additive Model Selection: Application to Electricity Demand Forecasting %A Keshav Das %A Julie Keisler %A Margaux Brégère %A Amaury Durand %B Proceedings of the Fourth International Conference on Automated Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Leman Akoglu %E Carola Doerr %E Jan N. van Rijn %E Roman Garnett %E Jacob R. Gardner %F pmlr-v293-das25a %I PMLR %P 23/1--19 %U https://proceedings.mlr.press/v293/das25a.html %V 293 %X 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, 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 to optimize them using automated Machine Learning. For this purpose, we define an efficient modeling of the search space (namely, the space of the GAM formulae and adaptation parameters) as well as mutation and crossover operators on this space and apply an evolutionary algorithm to select the best parameters. We evaluate our method on short-term French electricity demand forecasting which demonstrates the relevance of the approach.
APA
Das, K., Keisler, J., Brégère, M. & Durand, A.. (2025). AutoML Algorithms for Online Generalized Additive Model Selection: Application to Electricity Demand Forecasting. Proceedings of the Fourth International Conference on Automated Machine Learning, in Proceedings of Machine Learning Research 293:23/1-19 Available from https://proceedings.mlr.press/v293/das25a.html.

Related Material