Contextualizing MLP-Mixers Spatiotemporally for Urban Data Forecast at Scale

TL;DR

This paper introduces NexuSQN, a simplified yet effective MLP-based model for large-scale spatiotemporal traffic data forecasting, demonstrating competitive performance and versatility across multiple domains and real-world applications.

Contribution

The paper adapts the MLP-Mixer architecture for spatiotemporal data, proposing ST-contextualization to improve pattern distinction, and validates its effectiveness on traffic, energy, and environment datasets.

Findings

NexuSQN rivals state-of-the-art baselines in traffic forecasting.

The model generalizes well across different domains.

Successful deployment in real-world megacity congestion prediction.

Abstract

Spatiotemporal traffic data (STTD) displays complex correlational structures. Extensive advanced techniques have been designed to capture these structures for effective forecasting. However, because STTD is often massive in scale, practitioners need to strike a balance between effectiveness and efficiency using computationally efficient models. An alternative paradigm based on multilayer perceptron (MLP) called MLP-Mixer has the potential for both simplicity and effectiveness. Taking inspiration from its success in other domains, we propose an adapted version, named NexuSQN, for STTD forecast at scale. We first identify the challenges faced when directly applying MLP-Mixers as seriesand window-wise multivaluedness. To distinguish between spatial and temporal patterns, the concept of ST-contextualization is then proposed. Our results surprisingly show that this simple-yeteffective solution can rival SOTA baselines when tested on several traffic benchmarks. Furthermore, NexuSQN has demonstrated its versatility across different domains, including energy and environment data, and has been deployed in a collaborative project with Baidu to predict congestion in megacities like Beijing and Shanghai. Our findings contribute to the exploration of simple-yet-effective models for real-world STTD forecasting.