Robust Spatiotemporal Traffic Forecasting with Reinforced Dynamic Adversarial Training
This work addresses the critical problem of adversarial robustness in spatiotemporal traffic forecasting for Intelligent Transportation Systems, representing an incremental improvement over existing methods.
The paper tackles the vulnerability of machine learning-based traffic forecasting models to adversarial attacks by proposing a novel framework that incorporates adversarial training, using reinforcement learning for optimal node selection and self-knowledge distillation to address dynamic attacks and forgetting issues, achieving enhanced robustness on real-world datasets.
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.