Spatial-Temporal-Decoupled Masked Pre-training for Spatiotemporal Forecasting

TL;DR

This paper introduces STD-MAE, a self-supervised pre-training framework with decoupled autoencoders for improved spatiotemporal forecasting, addressing input length limitations and heterogeneity challenges.

Contribution

It proposes a novel decoupled masked autoencoder approach for spatiotemporal series reconstruction, enhancing downstream forecasting performance.

Findings

Achieves state-of-the-art results on six benchmark datasets.

Effectively captures complex spatiotemporal heterogeneity.

Improves prediction accuracy over existing methods.

Abstract

Spatiotemporal forecasting techniques are significant for various domains such as transportation, energy, and weather. Accurate prediction of spatiotemporal series remains challenging due to the complex spatiotemporal heterogeneity. In particular, current end-to-end models are limited by input length and thus often fall into spatiotemporal mirage, i.e., similar input time series followed by dissimilar future values and vice versa. To address these problems, we propose a novel self-supervised pre-training framework Spatial-Temporal-Decoupled Masked Pre-training (STD-MAE) that employs two decoupled masked autoencoders to reconstruct spatiotemporal series along the spatial and temporal dimensions. Rich-context representations learned through such reconstruction could be seamlessly integrated by downstream predictors with arbitrary architectures to augment their performances. A series of quantitative and qualitative evaluations on six widely used benchmarks (PEMS03, PEMS04, PEMS07, PEMS08, METR-LA, and PEMS-BAY) are conducted to validate the state-of-the-art performance of STD-MAE. Codes are available at https://github.com/Jimmy-7664/STD-MAE.