Accurate traffic-flow forecasting remains challenged by abrupt and irregular states even after dominant periodic patterns are captured. Existing predictors model the resulting difficult errors implicitly through their parameters and cannot explicitly reuse specific historical errors at inference. We find that multi-hor...
Qian-Xin Xie, Jin-Feng Xu, Yu-Chen Lu et al.· Mathematics· 0 citations
RIFT-STGNN, a Robust Interleaved Frequency–Trend Spatio-Temporal Graph Neural Network for multi-step traffic flow forecasting, follows a coordinated information flow and shows competitive numerical performance relative to selected literature-reported baselines under the 12-step setting.
Qian-Xin Xie, Jin-Feng Xu, Yuchen Lu et al.· Mathematics· 1 citation
F$^2$STNet is proposed, a federated forecasting framework that combines truncated graph-Fourier features, a lightweight diagonal state-space temporal encoder, graph convolution, and Fairness-aware Federated Aggregation (FFA).
Jia-Yi Zhang, Jin-Feng Xu, Hewei Wang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.