Open access
2026
Physics-Informed Generative Modeling for Sparse OD Matrix Forecasting: A Dual-Head VAE-GMM Approach With a Single Time Step Input
This study proposes a novel, compact, physics-informed generative framework capable of forecasting transit demands using only a single historical time step as input, which successfully mitigates zero-inflation and performs significantly better than traditional parametric and deep-learning baselines.
Mark Mpabulungi, Chang-Hun Kang, Keemin Sohn et al.
· IEEE Access · 0 citations