This paper investigates traffic forecasting, which attempts to forecast the future state of traffic based on historical situations. This problem has received ever-increasing attention in various scenarios and facilitated the development of numerous downstream applications such as urban planning and transportation management. However, the efficacy of existing methods remains sub-optimal due to their tendency to model temporal and spatial relationships independently, thereby inadequately accounting for complex high-order interactions of both worlds. Moreover, the diversity of transitional patterns in traffic forecasting makes them challenging to capture for existing approaches, warranting a deeper exploration of their diversity. Toward this end, this paper proposes njoint Spati-Tempora graph neural network (abbreviated as COOL), which models heterogeneous graphs from prior and posterior information to conjointly capture high-order spatio-temporal relationships. On the one hand, heterogeneous graphs connecting sequential observation are constructed to extract composite spatio-temporal relationships via prior message passing. On the other hand, we model dynamic relationships using constructed affinity and penalty graphs, which guide posterior message passing to incorporate complementary semantic information into node representations. Moreover, to capture diverse transitional properties to enhance traffic forecasting, we propose a conjoint self-attention decoder that models diverse temporal patterns from both multi-rank and multi-scale views. Experimental results on four popular benchmark datasets demonstrate that our proposed COOL provides state-of-the-art performance compared with the competitive baselines.

中文翻译：

---

COOL：用于交通预测的时空图神经网络的联合视角

本文研究了交通预测，试图根据历史情况预测未来的交通状态。这个问题在各种场景中受到越来越多的关注，并促进了城市规划、交通管理等众多下游应用的发展。然而，现有方法的功效仍然不是最佳的，因为它们倾向于独立地对时间和空间关系进行建模，从而不足以解释两个世界的复杂高阶相互作用。此外，交通预测中过渡模式的多样性使得现有方法难以捕获它们，因此需要对其多样性进行更深入的探索。为此，本文提出了njoint Spati-Tempora图神经网络（简称COOL），它根据先验和后验信息对异构图进行建模，以联合捕获高阶时空关系。一方面，构建连接顺序观察的异构图，以通过先验消息传递提取复合时空关系。另一方面，我们使用构造的亲和力和惩罚图对动态关系进行建模，这指导后验消息传递将补充语义信息合并到节点表示中。此外，为了捕获不同的过渡属性以增强流量预测，我们提出了一种联合自注意力解码器，它可以从多等级和多尺度视图对不同的时间模式进行建模。四个流行基准数据集的实验结果表明，与竞争基准相比，我们提出的 COOL 提供了最先进的性能。