Original title: Self-Supervised Deconfounding Against Spatio-Temporal Shifts: Theory and Modeling
Authors: Jiahao Ji, Wentao Zhang, Jingyuan Wang, Yue He, Chao Huang
In this article, the focus is on predicting traffic patterns in cities, crucial for efficient urban travel and sustainable growth. However, these predictions often stumble when faced with changes in traffic flow due to factors like time and location differences.
Original article: https://arxiv.org/abs/2311.12472