Linear Attention is Enough in Spatial-Temporal Forecasting
This addresses spatial-temporal forecasting challenges in traffic for machine learning applications, offering a novel approach to capture dynamic topology and integrated relationships, though it builds incrementally on Transformer architectures.
The paper tackles the problem of spatial-temporal forecasting, particularly in traffic, by proposing STformer, a method that treats nodes at different time steps as tokens in a Transformer to learn complex patterns, achieving state-of-the-art performance with affordable computational cost, and introduces NSTformer with linear complexity that slightly outperforms it in some cases.
As the most representative scenario of spatial-temporal forecasting tasks, the traffic forecasting task attracted numerous attention from machine learning community due to its intricate correlation both in space and time dimension. Existing methods often treat road networks over time as spatial-temporal graphs, addressing spatial and temporal representations independently. However, these approaches struggle to capture the dynamic topology of road networks, encounter issues with message passing mechanisms and over-smoothing, and face challenges in learning spatial and temporal relationships separately. To address these limitations, we propose treating nodes in road networks at different time steps as independent spatial-temporal tokens and feeding them into a vanilla Transformer to learn complex spatial-temporal patterns, design \textbf{STformer} achieving SOTA. Given its quadratic complexity, we introduce a variant \textbf{NSTformer} based on Nystr$\ddot{o}$m method to approximate self-attention with linear complexity but even slightly better than former in a few cases astonishingly. Extensive experimental results on traffic datasets demonstrate that the proposed method achieves state-of-the-art performance at an affordable computational cost. Our code is available at \href{https://github.com/XinyuNing/STformer-and-NSTformer}{https://github.com/XinyuNing/STformer-and-NSTformer}.