Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2410.15217 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912610213953536 |
|---|---|
| author | Gunasekaran, Skye Kembay, Assel Ladret, Hugo Zhu, Rui-Jie Perrinet, Laurent Kavehei, Omid Eshraghian, Jason |
| author_facet | Gunasekaran, Skye Kembay, Assel Ladret, Hugo Zhu, Rui-Jie Perrinet, Laurent Kavehei, Omid Eshraghian, Jason |
| contents | Accurate time-series forecasting is crucial in various scientific and industrial domains, yet deep learning models often struggle to capture long-term dependencies and adapt to data distribution shifts over time. We introduce Future-Guided Learning, an approach that enhances time-series event forecasting through a dynamic feedback mechanism inspired by predictive coding. Our method involves two models: a detection model that analyzes future data to identify critical events and a forecasting model that predicts these events based on current data. When discrepancies occur between the forecasting and detection models, a more significant update is applied to the forecasting model, effectively minimizing surprise, allowing the forecasting model to dynamically adjust its parameters. We validate our approach on a variety of tasks, demonstrating a 44.8% increase in AUC-ROC for seizure prediction using EEG data, and a 23.4% reduction in MSE for forecasting in nonlinear dynamical systems (outlier excluded).By incorporating a predictive feedback mechanism, Future-Guided Learning advances how deep learning is applied to time-series forecasting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_15217 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Predictive Approach To Enhance Time-Series Forecasting Gunasekaran, Skye Kembay, Assel Ladret, Hugo Zhu, Rui-Jie Perrinet, Laurent Kavehei, Omid Eshraghian, Jason Machine Learning Accurate time-series forecasting is crucial in various scientific and industrial domains, yet deep learning models often struggle to capture long-term dependencies and adapt to data distribution shifts over time. We introduce Future-Guided Learning, an approach that enhances time-series event forecasting through a dynamic feedback mechanism inspired by predictive coding. Our method involves two models: a detection model that analyzes future data to identify critical events and a forecasting model that predicts these events based on current data. When discrepancies occur between the forecasting and detection models, a more significant update is applied to the forecasting model, effectively minimizing surprise, allowing the forecasting model to dynamically adjust its parameters. We validate our approach on a variety of tasks, demonstrating a 44.8% increase in AUC-ROC for seizure prediction using EEG data, and a 23.4% reduction in MSE for forecasting in nonlinear dynamical systems (outlier excluded).By incorporating a predictive feedback mechanism, Future-Guided Learning advances how deep learning is applied to time-series forecasting. |
| title | A Predictive Approach To Enhance Time-Series Forecasting |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2410.15217 |