TESTAM: A Time-Enhanced Spatio-Temporal Attention Model with Mixture of Experts

arXiv:2403.02600v1 Announce Type: new
Abstract: Accurate traffic forecasting is challenging due to the complex dependency on road networks, various types of roads, and the abrupt speed change due to the events. Recent works mainly focus on dynamic spatial modeling with adaptive graph embedding or graph attention having less consideration for temporal characteristics and in-situ modeling. In this paper, we propose a novel deep learning model named TESTAM, which individually models recurring and non-recurring traffic patterns by a mixture-of-experts model with …

arxiv attention cs.lg cs.si experts mixture of experts temporal type