Reservoir Computing for Short High-Dimensional Time Series: an Application to SARS-CoV-2 Hospitalization Forecast

Thomas Ferté, Dan Dutartre, Boris P Hejblum, Romain Griffier, Vianney Jouhet, Rodolphe Thiébaut, Pierrick Legrand, Xavier Hinaut
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:13570-13591, 2024.

Abstract

In this work, we aimed at forecasting the number of SARS-CoV-2 hospitalized patients at 14 days to help anticipate the bed requirements of a large scale hospital using public data and electronic health records data. Previous attempts led to mitigated performance in this high-dimension setting; we introduce a novel approach to time series forecasting by providing an alternative to conventional methods to deal with high number of potential features of interest (409 predictors). We integrate Reservoir Computing (RC) with feature selection using a genetic algorithm (GA) to gather optimal non-linear combinations of inputs to improve prediction in sample-efficient context. We illustrate that the RC-GA combination exhibits excellent performance in forecasting SARS-CoV-2 hospitalizations. This approach outperformed the use of RC alone and other conventional methods: LSTM, Transformers, Elastic-Net, XGBoost. Notably, this work marks the pioneering use of RC (along with GA) in the realm of short and high-dimensional time series, positioning it as a competitive and innovative approach in comparison to standard methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-ferte24a, title = {Reservoir Computing for Short High-Dimensional Time Series: an Application to {SARS}-{C}o{V}-2 Hospitalization Forecast}, author = {Fert\'{e}, Thomas and Dutartre, Dan and Hejblum, Boris P and Griffier, Romain and Jouhet, Vianney and Thi\'{e}baut, Rodolphe and Legrand, Pierrick and Hinaut, Xavier}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {13570--13591}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/ferte24a/ferte24a.pdf}, url = {https://proceedings.mlr.press/v235/ferte24a.html}, abstract = {In this work, we aimed at forecasting the number of SARS-CoV-2 hospitalized patients at 14 days to help anticipate the bed requirements of a large scale hospital using public data and electronic health records data. Previous attempts led to mitigated performance in this high-dimension setting; we introduce a novel approach to time series forecasting by providing an alternative to conventional methods to deal with high number of potential features of interest (409 predictors). We integrate Reservoir Computing (RC) with feature selection using a genetic algorithm (GA) to gather optimal non-linear combinations of inputs to improve prediction in sample-efficient context. We illustrate that the RC-GA combination exhibits excellent performance in forecasting SARS-CoV-2 hospitalizations. This approach outperformed the use of RC alone and other conventional methods: LSTM, Transformers, Elastic-Net, XGBoost. Notably, this work marks the pioneering use of RC (along with GA) in the realm of short and high-dimensional time series, positioning it as a competitive and innovative approach in comparison to standard methods.} }
Endnote
%0 Conference Paper %T Reservoir Computing for Short High-Dimensional Time Series: an Application to SARS-CoV-2 Hospitalization Forecast %A Thomas Ferté %A Dan Dutartre %A Boris P Hejblum %A Romain Griffier %A Vianney Jouhet %A Rodolphe Thiébaut %A Pierrick Legrand %A Xavier Hinaut %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-ferte24a %I PMLR %P 13570--13591 %U https://proceedings.mlr.press/v235/ferte24a.html %V 235 %X In this work, we aimed at forecasting the number of SARS-CoV-2 hospitalized patients at 14 days to help anticipate the bed requirements of a large scale hospital using public data and electronic health records data. Previous attempts led to mitigated performance in this high-dimension setting; we introduce a novel approach to time series forecasting by providing an alternative to conventional methods to deal with high number of potential features of interest (409 predictors). We integrate Reservoir Computing (RC) with feature selection using a genetic algorithm (GA) to gather optimal non-linear combinations of inputs to improve prediction in sample-efficient context. We illustrate that the RC-GA combination exhibits excellent performance in forecasting SARS-CoV-2 hospitalizations. This approach outperformed the use of RC alone and other conventional methods: LSTM, Transformers, Elastic-Net, XGBoost. Notably, this work marks the pioneering use of RC (along with GA) in the realm of short and high-dimensional time series, positioning it as a competitive and innovative approach in comparison to standard methods.
APA
Ferté, T., Dutartre, D., Hejblum, B.P., Griffier, R., Jouhet, V., Thiébaut, R., Legrand, P. & Hinaut, X.. (2024). Reservoir Computing for Short High-Dimensional Time Series: an Application to SARS-CoV-2 Hospitalization Forecast. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:13570-13591 Available from https://proceedings.mlr.press/v235/ferte24a.html.

Related Material