DRAN: A Distribution and Relation Adaptive Network for Spatio-temporal Forecasting
DRAN addresses spatio-temporal forecasting challenges by integrating two novel modules: a Spatial Factor Learner (SFL) for distribution adaptation and a Dynamic-Static Fusion Learner (DSFL) to model evolving spatial dependencies. DRAN outperforms existing methods on six datasets by adaptively fusing static and dynamic spatial components, leading to superior accuracy in point predictions and distributional metrics.