HSTMixer: A Hierarchical MLP-Mixer for Large-Scale Traffic Forecasting

TL;DR

HSTMixer is a novel hierarchical MLP-based framework designed for efficient large-scale traffic forecasting, capturing multi-resolution spatiotemporal patterns and regional semantics to outperform existing models.

Contribution

The paper introduces HSTMixer, a hierarchical MLP-Mixer architecture with adaptive regional mixing for scalable and accurate traffic prediction.

Findings

Achieves state-of-the-art accuracy on four large-scale datasets.

Demonstrates superior computational efficiency compared to existing models.

Effectively captures multi-resolution spatiotemporal features.

Abstract

Traffic forecasting task is significant to modern urban management. Recently, there is growing attention on large-scale forecasting, as it better reflects the complexity of real-world traffic networks. However, existing models often exhibit quadratic computational complexity, making them impractical for large-scale real-world scenarios. In this paper, we propose a novel framework, Hierarchical Spatio-Temporal Mixer (HSTMixer), which leverages an all-MLP architecture for efficient and effective large-scale traffic forecasting. HSTMixer employs a hierarchical spatiotemporal mixing block to extract multi-resolution features through bottom-up aggregation and top-down propagation. Furthermore, an adaptive region mixer generates transformation matrices based on regional semantics, enabling our model to dynamically capture evolving spatiotemporal patterns for different regions. Extensive experiments conducted on four large-scale real-world datasets demonstrate that the proposed method not only achieves state-of-the-art performance but also exhibits competitive computational efficiency.