Research map / Graph learning / Knowledge and temporal graphs
Micro-Macro Coupled Koopman Modeling on Graph for Traffic Flow Prediction
Bairan Xiang, Chenguang Zhao, Huan Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 changed | Where | Measured | Effect |
|---|---|---|---|
| add Intent Discriminator (ablation) | MMCKM · HighD (1s) | RMSE: 0.74 -> 0.29 | Improved |
| add Koopman control module (ablation) | MMCKM · HighD (5s) | RMSE: 3.46 -> 2.73 | Improved |
| add both Intent Discriminator and Koopman control (ablation) | MMCKM · HighD (5s) | RMSE: 4.62 -> 2.73 | Improved |
| add diffusion term to advection-only model (ablation) | MMCKM macro module · NGSIM macro density (5s) | RMSE (%): 14.1% -> 9.5% | Improved |
| set KDE bandwidth to 25 m instead of 10 m (ablation) | MMCKM macro module · NGSIM macro density (5s) | RMSE (%): 14.0% -> 9.5% | Improved |
| use KDE bandwidth 10 m with diffusion (vs advection-only) (ablation) | LC · NGSIM macro density (1s) | RMSE (%): 6.1% -> 6.4% | Hurt |
| use 1.0s operator instead of 0.1s | MMCKM · NGSIM trajectory (5s) | RMSE: 4.65 -> 2.93 | Improved |
| use 1.0s operator instead of 0.1s | MMCKM · NGSIM trajectory (1s) | RMSE: 0.33 -> 0.54 | Hurt |
| increase operator interval from 0.04s to 0.4s (ablation) | MMCKM · HighD | ADE: 2.84 -> 1.65 | Improved |
| increase operator interval from 0.4s to 1s (ablation) | MMCKM · HighD | ADE: 1.65 -> 2.90 | Hurt |
| use history-free MMCKM (1.0s) instead of CV history-free reference | MMCKM · NGSIM trajectory (5s) | RMSE: 5.62 -> 2.93 | Improved |
| increase KDE bandwidth from 25 m to 50 m (ablation) | LC · NGSIM macro density (5s) | RMSE (%): 9.5% -> 12.6% | Hurt |
Closest papers by meaning
- Deep-Koopman-KANDy: Dictionary Discovery for Deep-Koopman Operators with Kolmogorov-Arnold Networks for Dynamics (NeurIPS 2026 · similarity 0.84)
- NEST: A Neuromodulated Small-world Hypergraph Trajectory Prediction Model for Autonomous Driving (AAAI 2025 · similarity 0.84)
- Flow Matching-Based Autonomous Driving Planning with Advanced Interactive Behavior Modeling (NeurIPS 2025 · similarity 0.84)
- LASIL: Learner-Aware Supervised Imitation Learning For Long-term Microscopic Traffic Simulation (CVPR 2024 · similarity 0.84)
- Density-Adaptive Model Based on Motif Matrix for Multi-Agent Trajectory Prediction (CVPR 2024 · similarity 0.83)
- Trajeglish: Traffic Modeling as Next-Token Prediction (ICLR 2024 · similarity 0.83)
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.
