用因果分析提升事故引发拥堵的预测精度。
CASTANET: Causality-Aware Spatio-Temporal Adversarial Network Using Traffic Incident Effects
- 结合图神经网络与因果效应估计，显式建模事故影响。
- 在东京真实数据上，整体RMSE降低4.0%，事故场景下降10.1%。
- 适合交通系统优化、智能导航等需要精准应急预测的场景。
预测由突发事故（如车祸、道路损坏）引起的非周期性交通拥堵，对高级智能交通系统至关重要。然而，由于事故稀疏、时空分布不均且影响依赖交通背景，其预测难度大。现有深度学习方法虽提升了周期性拥堵预测性能，但在非周期性场景中仍受限，因事故记录未被显式利用且存在严重时空偏倚。为此，我们提出CASTANET，融合时空图神经网络与因果处理效应估计，有效利用事故记录并缓解选择偏倚。基于东京真实交通数据与事故记录（视为事故）的实验表明，相比最优基线，CASTANET总体RMSE降低4.0%，在事故条件评估中降低10.1%，极端拥堵下最高提升达14.55%。
原文摘要 · Abstract (English)
Predicting non-periodic traffic congestion caused by sudden incidents (e.g., accidents and road damage) is crucial for advanced intelligent transportation systems. However, incident-driven congestion is difficult to forecast because incidents are extremely sparse, occur at specific times and locations, and have heterogeneous impacts depending on the traffic context. While recent deep learning approaches have significantly improved periodic traffic forecasting, their performance on non-periodic congestion remains limited, partly because incident records are not explicitly incorporated and their occurrence is strongly biased in space and time. To address these challenges, we propose CASTANET, which integrates spatio-temporal graph neural networks and causal treatment effect estimation to utilize incident records while mitigating selection bias. Experiments on real-world traffic data and accident records from Tokyo, which we treat as incidents, show that CASTANET reduces RMSE by 4.0% overall compared to the best baseline and by 10.1% on incident-conditioned evaluation, with gains reaching 14.55% under severe congestion.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。