Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

FDN: Interpretable Spatiotemporal Forecasting with Future Decomposition Networks

About

Spatiotemporal systems comprise a collection of spatially distributed yet interdependent entities each generating unique dynamic signals. Highly sophisticated methods have been proposed in recent years delivering state-of-the-art (SOTA) forecasts but few have focused on interpretability. To address this, we propose the Future Decomposition Network (FDN), a novel forecast model capable of (a) providing interpretable predictions through classification (b) revealing latent activity patterns in the target time-series and (c) delivering forecasts competitive with SOTA methods at a fraction of their memory and runtime cost. We conduct comprehensive analyses on FDN for multiple datasets from hydrologic, traffic, and energy systems, demonstrating its improved accuracy and interpretability.

Nicholas Majeske, Ariful Azad• 2026

Related benchmarks

TaskDatasetResultRank
Spatiotemporal forecastingWabash River (test)
MAE3.127
36
Spatiotemporal forecastingE-PEMS-BAY (test)
MAE0.937
36
ForecastingSolar-Energy (test)
MAE0.238
27
Spatiotemporal forecastingSolar-Energy (test)
MAE0.238
27
ForecastingE-PEMS-BAY Horizon 6
MAE1.475
12
ForecastingE-PEMS-BAY Horizon 12
MAE1.792
12
Time Series ForecastingWabash River Basin Horizon 4
MAE6.624
12
Time Series ForecastingWabash River Basin Horizon 7
MAE9.122
12
ForecastingE-PEMS-BAY Horizon 1
MAE0.937
12
Time Series ForecastingWabash River Basin Horizon 1
MAE3.127
12
Showing 10 of 10 rows

Other info

Follow for update