atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

STDEN: Towards Physics-Guided Neural Networks for Traffic Flow Prediction

AAAI 2022 · Probabilistic and scientific modeling › Molecules and proteins · approach: Spatiotemporal forecasting and imputation · 106 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use STDEN's physics-guided potential-energy-field differential equation instead of the strongest baselineTraffic flow prediction · GT-221, 15 minMAE: 0.962 -> 0.865Helped
Use STDEN's physics-guided potential-energy-field differential equation instead of the strongest baselineTraffic flow prediction · WRS-393, 15 minMAE: 0.792 -> 0.730Helped
Use STDEN's physics-guided potential-energy-field differential equation instead of the strongest baselineTraffic flow prediction · ZGC-564, 15 minMAE: 0.708 -> 0.623Helped
Replace the potential-energy-field DE with a fully connected network (unknown physics) (ablation)STDEN · All three datasetsMAEHurt
Remove the energy volume factor φ from the potential-energy-field DE (ablation)STDEN · All three datasetsMAEHurt
Increase the ODE solver's number of function evaluationsSTDEN · GT-221, WRS-393, and ZGC-564MAEHelped

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy