Self-Supervised State Space Model for Real-Time Traffic Accident Prediction Using eKAN Networks

TL;DR

The paper introduces SSL-eKamba, a self-supervised framework that enhances real-time traffic accident prediction by improving generalization through auxiliary tasks and increasing efficiency with a redesigned network architecture.

Contribution

It presents a novel self-supervised approach with auxiliary tasks and an efficient network design for accurate, scalable, and real-time traffic accident prediction.

Findings

Outperforms state-of-the-art baselines on real-world datasets

Enhances generalization with self-supervised auxiliary tasks

Achieves real-time performance through efficient network redesign

Abstract

Accurate prediction of traffic accidents across different times and regions is vital for public safety. However, existing methods face two key challenges: 1) Generalization: Current models rely heavily on manually constructed multi-view structures, like POI distributions and road network densities, which are labor-intensive and difficult to scale across cities. 2) Real-Time Performance: While some methods improve accuracy with complex architectures, they often incur high computational costs, limiting their real-time applicability. To address these challenges, we propose SSL-eKamba, an efficient self-supervised framework for traffic accident prediction. To enhance generalization, we design two self-supervised auxiliary tasks that adaptively improve traffic pattern representation through spatiotemporal discrepancy awareness. For real-time performance, we introduce eKamba, an efficient model that redesigns the Kolmogorov-Arnold Network (KAN) architecture. This involves using learnable univariate functions for input activation and applying a selective mechanism (Selective SSM) to capture multi-variate correlations, thereby improving computational efficiency. Extensive experiments on two real-world datasets demonstrate that SSL-eKamba consistently outperforms state-of-the-art baselines. This framework may also offer new insights for other spatiotemporal tasks. Our source code is publicly available at http://github.com/KevinT618/SSL-eKamba.