Reversible Residual Normalization Alleviates Spatio-Temporal Distribution Shift

TL;DR

This paper introduces Reversible Residual Normalization, a graph-based invertible framework that adaptively normalizes spatio-temporal data, effectively mitigating distribution shifts in deep forecasting models.

Contribution

It proposes a novel invertible normalization method combining graph convolutional operations with spectral constraints to handle complex spatio-temporal distribution shifts.

Findings

Improves forecasting accuracy under distribution shift.

Maintains reversibility for better model interpretability.

Captures complex spatio-temporal relationships effectively.

Abstract

Distribution shift severely degrades the performance of deep forecasting models. While this issue is well-studied for individual time series, it remains a significant challenge in the spatio-temporal domain. Effective solutions like instance normalization and its variants can mitigate temporal shifts by standardizing statistics. However, distribution shift on a graph is far more complex, involving not only the drift of individual node series but also heterogeneity across the spatial network where different nodes exhibit distinct statistical properties. To tackle this problem, we propose Reversible Residual Normalization (RRN), a novel framework that performs spatially-aware invertible transformations to address distribution shift in both spatial and temporal dimensions. Our approach integrates graph convolutional operations within invertible residual blocks, enabling adaptive normalization that respects the underlying graph structure while maintaining reversibility. By combining Center Normalization with spectral-constrained graph neural networks, our method captures and normalizes complex Spatio-Temporal relationships in a data-driven manner. The bidirectional nature of our framework allows models to learn in a normalized latent space and recover original distributional properties through inverse transformation, offering a robust and model-agnostic solution for forecasting on dynamic spatio-temporal systems.