A General ReLearner: Empowering Spatiotemporal Prediction by Re-learning Input-label Residual

TL;DR

This paper introduces ReLearner, a bidirectional learning module that improves spatiotemporal prediction accuracy by explicitly modeling residuals between inputs and labels, addressing limitations of traditional unidirectional models.

Contribution

It proposes the Spatiotemporal Residual Theorem and a universal ReLearner module that enhances existing neural networks with bidirectional residual learning capabilities.

Findings

ReLearner significantly improves prediction accuracy across multiple datasets.

The module effectively disentangles and smooths residuals, leading to more stable training.

Experiments show consistent performance gains on 14 backbone models.

Abstract

Prevailing spatiotemporal prediction models typically operate under a forward (unidirectional) learning paradigm, in which models extract spatiotemporal features from historical observation input and map them to target spatiotemporal space for future forecasting (label). However, these models frequently exhibit suboptimal performance when spatiotemporal discrepancies exist between inputs and labels, for instance, when nodes with similar time-series inputs manifest distinct future labels, or vice versa. To address this limitation, we propose explicitly incorporating label features during the training phase. Specifically, we introduce the Spatiotemporal Residual Theorem, which generalizes the conventional unidirectional spatiotemporal prediction paradigm into a bidirectional learning framework. Building upon this theoretical foundation, we design an universal module, termed ReLearner, which seamlessly augments Spatiotemporal Neural Networks (STNNs) with a bidirectional learning capability via an auxiliary inverse learning process. In this process, the model relearns the spatiotemporal feature residuals between input data and future data. The proposed ReLearner comprises two critical components: (1) a Residual Learning Module, designed to effectively disentangle spatiotemporal feature discrepancies between input and label representations; and (2) a Residual Smoothing Module, employed to smooth residual terms and facilitate stable convergence. Extensive experiments conducted on 11 real-world datasets across 14 backbone models demonstrate that ReLearner significantly enhances the predictive performance of existing STNNs.Our code is available on GitHub.