A graph neural network that learns one dependency graph per decomposed time-series component improves long-term urban forecasts by a few percent over baselines.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Forecasting Multivariate Urban Data via Decomposition and Spatio-Temporal Graph Analysis
A graph neural network that learns one dependency graph per decomposed time-series component improves long-term urban forecasts by a few percent over baselines.