Robust Spatiotemporal Traffic Forecasting with Reinforced Dynamic Adversarial Training

TL;DR

This paper introduces a reinforcement learning-based adversarial training framework to improve the robustness of spatiotemporal traffic forecasting models against dynamic adversarial attacks, addressing limitations of static domain methods.

Contribution

It proposes a novel RL-based node selection strategy and a self-knowledge distillation module tailored for dynamic adversarial defense in traffic forecasting.

Findings

Enhanced robustness against adversarial attacks on traffic data

Outperforms baseline models on real-world datasets

Reduces overfitting during adversarial training

Abstract

Machine learning-based forecasting models are commonly used in Intelligent Transportation Systems (ITS) to predict traffic patterns and provide city-wide services. However, most of the existing models are susceptible to adversarial attacks, which can lead to inaccurate predictions and negative consequences such as congestion and delays. Therefore, improving the adversarial robustness of these models is crucial for ITS. In this paper, we propose a novel framework for incorporating adversarial training into spatiotemporal traffic forecasting tasks. We demonstrate that traditional adversarial training methods designated for static domains cannot be directly applied to traffic forecasting tasks, as they fail to effectively defend against dynamic adversarial attacks. Then, we propose a reinforcement learning-based method to learn the optimal node selection strategy for adversarial examples, which simultaneously strengthens the dynamic attack defense capability and reduces the model overfitting. Additionally, we introduce a self-knowledge distillation regularization module to overcome the "forgetting issue" caused by continuously changing adversarial nodes during training. We evaluate our approach on two real-world traffic datasets and demonstrate its superiority over other baselines. Our method effectively enhances the adversarial robustness of spatiotemporal traffic forecasting models. The source code for our framework is available at https://github.com/usail-hkust/RDAT.