GAS-PFN: Graph-Aware Spatiotemporal Prior-Data Fitted Networks for Zero-Shot Traffic Forecasting

📄 arXiv: 2610.37598v1 📥 PDF

作者: Joseph Natter, Long Vu, Yifan Zhang, Bobin Deng, Kun Suo, Rui Wu

分类: cs.LG, cs.AI, cs.NE

发布日期: 2026-10-27


💡 一句话要点

提出GAS-PFN以解决零样本交通预测中的图结构依赖问题

🎯 匹配领域: 支柱八:物理动画 (Physics-based Animation)

关键词: 交通预测 图神经网络 零样本学习 时空建模 深度学习

📋 核心要点

  1. 现有的基于图的交通预测模型在特定数据集上训练,导致在新网络上部署时成本高昂,缺乏灵活性。
  2. GAS-PFN通过引入图结构依赖,结合随机游走扩散和掩码条件推理,提供了一种零样本适应的解决方案。
  3. 在六个基准测试中,GAS-PFN相较于TimePFN和其他零样本预测模型显著降低了均方误差,显示出其有效性。

📝 摘要(中文)

交通预测需要建模复杂的时间动态和由道路拓扑施加的空间依赖性。现有的基于图的监督模型在特定数据集和传感器图上训练,导致在新网络上的部署成本高昂。为此,本文提出了图感知时空先验数据拟合网络(GAS-PFN),通过随机游走扩散将采样的道路拓扑融入合成任务先验,帮助模型在预训练阶段学习图耦合动态。其推理架构采用共享的Transformer骨干网,通过两个掩码条件的传递分别建模传感器特定的时间动态和相邻传感器之间的时空交互,最终整合其表示进行预测。实验结果表明,GAS-PFN在六个基准测试中相较于其前身TimePFN平均降低34.0%的均方误差(MSE),并在与四个行业开发的通用零样本预测模型的比较中平均降低27.7%的MSE。

🔬 方法详解

问题定义:本文旨在解决零样本交通预测中图结构依赖未被充分建模的问题。现有的预测模型通常在特定数据集上训练,缺乏对新网络的适应能力,导致部署成本高昂。

核心思路:GAS-PFN通过将采样的道路拓扑融入合成任务先验,利用随机游走扩散来学习图耦合动态,从而实现零样本适应。其推理架构设计为通过共享的Transformer骨干网进行两次掩码条件传递,分别处理传感器特定的时间动态和相邻传感器的时空交互。

技术框架:GAS-PFN的整体架构包括两个主要模块:一是通过随机游走扩散学习图结构的合成任务先验,二是采用掩码条件的Transformer推理架构,分别建模时间动态和时空交互,最后整合这些表示进行预测。

关键创新:GAS-PFN的创新在于其将图结构依赖显式地融入到先验数据拟合网络中,解决了现有方法在图结构建模上的不足,使得模型在新环境中无需参数更新即可进行有效预测。

关键设计:在设计上,GAS-PFN采用了共享的Transformer骨干网,使用掩码条件的两次传递来分别处理不同类型的动态,确保模型能够有效捕捉到传感器之间的时空关系。

🖼️ 关键图片

fig_0
fig_1

📊 实验亮点

在六个基准测试中,GAS-PFN相较于其前身TimePFN平均降低了34.0%的均方误差(MSE),并在与四个行业开发的通用零样本预测模型的比较中平均降低了27.7%的MSE,显示出显著的性能提升。

🎯 应用场景

GAS-PFN在智能交通系统、城市交通管理和自动驾驶等领域具有广泛的应用潜力。通过提高交通预测的准确性,能够帮助城市规划者和交通管理者优化交通流量,减少拥堵,提高出行效率。未来,该方法还可能扩展到其他需要时空预测的领域,如环境监测和公共安全。

📄 摘要(原文)

Traffic forecasting requires modeling complex temporal dynamics and the spatial dependencies imposed by road topology. Supervised graph-based models achieve strong performance, but are often trained for a particular dataset and sensor graph, making deployment to new or evolving networks costly. Prior-Data Fitted Networks (PFNs) offer a zero-shot alternative by learning in-context inference from synthetic task priors, which provide an explicit mechanism for encoding domain-specific assumptions. Existing forecasting PFNs, however, do not explicitly represent the graph-structured dependencies governing traffic dynamics. To combine graph-aware traffic modeling with zero-shot adaptation, we introduce the Graph-Aware Spatiotemporal Prior-Data Fitted Network (GAS-PFN). GAS-PFN incorporates sampled road topology into its synthetic task prior through random-walk diffusion, teaching the model to learn graph-coupled dynamics during pretraining. Its co-designed inference architecture uses two mask-conditioned passes through a shared Transformer backbone to separately model sensor-specific temporal dynamics and spatiotemporal interactions among neighboring sensors before integrating their representations for prediction. This design enables GAS-PFN to use target-network structure and context without parameter updates. Across six benchmarks at the 90-minute horizon, GAS-PFN reduces MSE by 34.0% on average relative to TimePFN, its closest PFN-based predecessor. It also reduces MSE by 27.7% on average relative to four recent industry-developed general-purpose zero-shot forecasters. In this comparison, all models receive 90 minutes of contiguous recent history, and any additional context is drawn from each dataset's chronological training split. These results indicate that aligning PFN priors and inference architectures with graph structure improves in-context learning for zero-shot traffic forecasting.