FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting
<p dir="ltr">Uncertainty quantification in spatiotemporal forecasting is crucial for decision-making. Quantile regression has been proposed as a computationally efficient and assumption-free approach for uncertainty quantification by generating prediction intervals. Yet, existing dee...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , |
| Published: |
2025
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1864513524397506560 |
|---|---|
| author | Shehel Yoosuf (23073535) |
| author2 | Hamza Baali (14603380) Abdesselam Bouzerdoum (17900021) |
| author2_role | author author |
| author_facet | Shehel Yoosuf (23073535) Hamza Baali (14603380) Abdesselam Bouzerdoum (17900021) |
| author_role | author |
| dc.creator.none.fl_str_mv | Shehel Yoosuf (23073535) Hamza Baali (14603380) Abdesselam Bouzerdoum (17900021) |
| dc.date.none.fl_str_mv | 2025-10-17T15:00:00Z |
| dc.identifier.none.fl_str_mv | 10.1109/ojcs.2025.3616224 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/journal_contribution/FiLM-SimVP_Scalable_Uncertainty_Quantification_in_Spatiotemporal_Forecasting/31169071 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Information and computing sciences Artificial intelligence Machine learning Quantile regression spatiotemporal forecasting probabilistic forecasting selfsupervised learning Computational modeling Predictive models Uncertainty Computer architecture Estimation Transformers Long short term memory Coordinate measuring machines |
| dc.title.none.fl_str_mv | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting |
| dc.type.none.fl_str_mv | Text Journal contribution info:eu-repo/semantics/publishedVersion text contribution to journal |
| description | <p dir="ltr">Uncertainty quantification in spatiotemporal forecasting is crucial for decision-making. Quantile regression has been proposed as a computationally efficient and assumption-free approach for uncertainty quantification by generating prediction intervals. Yet, existing deep learning based quantile regression methods struggle to scale efficiently with large datasets. This article introduces FiLM-SimVP, a novel architecture that combines Feature-wise Linear Modulation (FiLM) and recent advances in efficient 2D convolution-based attention mechanisms to improve spatiotemporal forecasting and interval prediction. The proposed approach enables adaptive modulation of spatiotemporal features by directly conditioning intermediate features on confidence levels, allowing the model to learn desired predictive intervals. Through extensive experimentation on three diverse datasets, namely TaxiBJ, Traffic4cast, and WeatherBench, we demonstrate that FiLM-SimVP consistently outperforms existing state-of-the-art (SOTA) methods in point estimation accuracy and interval quality metrics. The model achieves an improvement of 1.1% in Mean Absolute Error and 6% in Mean Squared Error compared to baseline approaches with only slightly increased parameters and computations. Additionally, FiLM-SimVP shows superior scalability when modeling multiple intervals, including the entire conditional distribution, effectively capturing the relationship between different confidence levels without requiring separate models.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: IEEE Open Journal of the Computer Society<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/ojcs.2025.3616224" target="_blank">https://dx.doi.org/10.1109/ojcs.2025.3616224</a></p> |
| eu_rights_str_mv | openAccess |
| id | Manara2_511c18abfaaa24963cf17fe09d9b12b4 |
| identifier_str_mv | 10.1109/ojcs.2025.3616224 |
| network_acronym_str | Manara2 |
| network_name_str | Manara2 |
| oai_identifier_str | oai:figshare.com:article/31169071 |
| publishDate | 2025 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal ForecastingShehel Yoosuf (23073535)Hamza Baali (14603380)Abdesselam Bouzerdoum (17900021)Information and computing sciencesArtificial intelligenceMachine learningQuantile regressionspatiotemporal forecastingprobabilistic forecastingselfsupervised learningComputational modelingPredictive modelsUncertaintyComputer architectureEstimationTransformersLong short term memoryCoordinate measuring machines<p dir="ltr">Uncertainty quantification in spatiotemporal forecasting is crucial for decision-making. Quantile regression has been proposed as a computationally efficient and assumption-free approach for uncertainty quantification by generating prediction intervals. Yet, existing deep learning based quantile regression methods struggle to scale efficiently with large datasets. This article introduces FiLM-SimVP, a novel architecture that combines Feature-wise Linear Modulation (FiLM) and recent advances in efficient 2D convolution-based attention mechanisms to improve spatiotemporal forecasting and interval prediction. The proposed approach enables adaptive modulation of spatiotemporal features by directly conditioning intermediate features on confidence levels, allowing the model to learn desired predictive intervals. Through extensive experimentation on three diverse datasets, namely TaxiBJ, Traffic4cast, and WeatherBench, we demonstrate that FiLM-SimVP consistently outperforms existing state-of-the-art (SOTA) methods in point estimation accuracy and interval quality metrics. The model achieves an improvement of 1.1% in Mean Absolute Error and 6% in Mean Squared Error compared to baseline approaches with only slightly increased parameters and computations. Additionally, FiLM-SimVP shows superior scalability when modeling multiple intervals, including the entire conditional distribution, effectively capturing the relationship between different confidence levels without requiring separate models.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: IEEE Open Journal of the Computer Society<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/ojcs.2025.3616224" target="_blank">https://dx.doi.org/10.1109/ojcs.2025.3616224</a></p>2025-10-17T15:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/ojcs.2025.3616224https://figshare.com/articles/journal_contribution/FiLM-SimVP_Scalable_Uncertainty_Quantification_in_Spatiotemporal_Forecasting/31169071CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/311690712025-10-17T15:00:00Z |
| spellingShingle | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting Shehel Yoosuf (23073535) Information and computing sciences Artificial intelligence Machine learning Quantile regression spatiotemporal forecasting probabilistic forecasting selfsupervised learning Computational modeling Predictive models Uncertainty Computer architecture Estimation Transformers Long short term memory Coordinate measuring machines |
| status_str | publishedVersion |
| title | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting |
| title_full | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting |
| title_fullStr | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting |
| title_full_unstemmed | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting |
| title_short | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting |
| title_sort | FiLM-SimVP: Scalable Uncertainty Quantification in Spatiotemporal Forecasting |
| topic | Information and computing sciences Artificial intelligence Machine learning Quantile regression spatiotemporal forecasting probabilistic forecasting selfsupervised learning Computational modeling Predictive models Uncertainty Computer architecture Estimation Transformers Long short term memory Coordinate measuring machines |