Vis enkel innførsel

dc.contributor.authorMarisca, Ivan
dc.contributor.authorAlippi, Cesare
dc.contributor.authorBianchi, Filippo Maria
dc.date.accessioned2024-06-10T10:53:57Z
dc.date.available2024-06-10T10:53:57Z
dc.date.issued2024-04-16
dc.description.abstractGiven a set of synchronous time series, each associated with a sensor-point in space and characterized by inter-series relationships, the problem of spatiotemporal forecasting consists of predicting future observations for each point. Spatiotemporal graph neural networks achieve striking results by representing the relationships across time series as a graph. Nonetheless, most existing methods rely on the often unrealistic assumption that inputs are always available and fail to capture hidden spatiotemporal dynamics when part of the data is missing. In this work, we tackle this problem through hierarchical spatiotemporal downsampling. The input time series are progressively coarsened over time and space, obtaining a pool of representations that capture heterogeneous temporal and spatial dynamics. Conditioned on observations and missing data patterns, such representations are combined by an interpretable attention mechanism to generate the forecasts. Our approach outperforms state-of-the-art methods on synthetic and real-world benchmarks under different missing data distributions, particularly in the presence of contiguous blocks of missing values.en_US
dc.identifier.citationMarisca, Alippi, Bianchi. Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling. Proceedings of Machine Learning Research (PMLR). 2024en_US
dc.identifier.cristinIDFRIDAID 2274624
dc.identifier.doi10.48550/arXiv.2402.10634
dc.identifier.issn2640-3498
dc.identifier.urihttps://hdl.handle.net/10037/33768
dc.language.isoengen_US
dc.publisherPMLRen_US
dc.relation.journalProceedings of Machine Learning Research (PMLR)
dc.rights.accessRightsopenAccessen_US
dc.rights.holderCopyright 2024 The Author(s)en_US
dc.rights.urihttps://creativecommons.org/licenses/by/4.0en_US
dc.rightsAttribution 4.0 International (CC BY 4.0)en_US
dc.titleGraph-based Forecasting with Missing Data through Spatiotemporal Downsamplingen_US
dc.type.versionsubmittedVersionen_US
dc.typeJournal articleen_US
dc.typeTidsskriftartikkelen_US


Tilhørende fil(er)

Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel

Attribution 4.0 International (CC BY 4.0)
Med mindre det står noe annet, er denne innførselens lisens beskrevet som Attribution 4.0 International (CC BY 4.0)