A Gated MLP Architecture for Learning Topological Dependencies in Spatio-Temporal Graphs

29 Jan 2024  ·  Yun Young Choi, Minho Lee, Sun Woo Park, Seunghwan Lee, Joohwan Ko ·

Graph Neural Networks (GNNs) and Transformer have been increasingly adopted to learn the complex vector representations of spatio-temporal graphs, capturing intricate spatio-temporal dependencies crucial for applications such as traffic datasets. Although many existing methods utilize multi-head attention mechanisms and message-passing neural networks (MPNNs) to capture both spatial and temporal relations, these approaches encode temporal and spatial relations independently, and reflect the graph's topological characteristics in a limited manner. In this work, we introduce the Cycle to Mixer (Cy2Mixer), a novel spatio-temporal GNN based on topological non-trivial invariants of spatio-temporal graphs with gated multi-layer perceptrons (gMLP). The Cy2Mixer is composed of three blocks based on MLPs: A message-passing block for encapsulating spatial information, a cycle message-passing block for enriching topological information through cyclic subgraphs, and a temporal block for capturing temporal properties. We bolster the effectiveness of Cy2Mixer with mathematical evidence emphasizing that our cycle message-passing block is capable of offering differentiated information to the deep learning model compared to the message-passing block. Furthermore, empirical evaluations substantiate the efficacy of the Cy2Mixer, demonstrating state-of-the-art performances across various traffic benchmark datasets.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Traffic Prediction PeMS04 Cy2Mixer 12 Steps MAE 18.14 # 2
Traffic Prediction PeMS07 Cy2Mixer MAE@1h 19.45 # 3
Traffic Prediction PeMS08 Cy2Mixer MAE@1h 13.53 # 3

Methods