SEP 10, 2026 · PREPRINT
When does a spectral prior help graph learning? Connectivity-loss estimation under road-network disruptions
arXiv
This is an unrefereed computational methods paper presenting a machine-learning approach to graph connectivity estimation, not a clinical or biomedical study; evidence rating does not apply.
Preprint: Not Peer ReviewedVan-Truong Le
Reported
Spatial-failure MAE improvement (…0.0391
95% hierarchical interval (spatia…0.0151–0.0662
Targeted-failure MAE improvement…0.0257