Research map / Graph learning / Knowledge and temporal graphs
Coupled Layer-wise Graph Convolution for Transportation Demand Prediction
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use CCRNN instead of Graph WaveNet | transportation demand prediction · NYC Taxi | RMSE: 13.0729 -> 9.5631; 26.85% lower | Helped |
| Use CCRNN instead of DCRNN | transportation demand prediction · NYC Citi Bike | RMSE: 3.2094 -> 2.8382 | Helped |
| Make the initial adjacency embeddings trainable (ablation) | CCRNN · NYC Citi Bike | RMSE: 3.2648 -> 2.8382 | Helped |
| Add layer-wise coupling (ablation) | CCRNN · NYC Citi Bike | RMSE: 3.0555 -> 2.8382 | Helped |
| Initialize adjacency embeddings from the learned graph rather than randomly (ablation) | CCRNN · NYC Citi Bike | PCC: 0.0204 -> 0.7934 | Helped |
| Initialize adjacency from station distances instead of the proposed initialization (ablation) | CCRNN · NYC Citi Bike | RMSE: 2.8382 -> 2.9718 | Hurt |
| Initialize adjacency from demand-series PCC instead of the proposed initialization (ablation) | CCRNN · NYC Citi Bike | RMSE: 2.8382 -> 2.9096 | Hurt |
| Make the initial adjacency embeddings trainable (ablation) | CCRNN · NYC Taxi | RMSE: 12.0516 -> 9.5631 | Helped |
| Add layer-wise coupling (ablation) | CCRNN · NYC Taxi | RMSE: 10.1864 -> 9.5631 | Helped |
| Initialize adjacency embeddings from the learned graph rather than randomly (ablation) | CCRNN · NYC Taxi | PCC: 0.0065 -> 0.9648 | Helped |
| Initialize adjacency from station distances instead of the proposed initialization (ablation) | CCRNN · NYC Taxi | RMSE: 9.5631 -> 10.6882 | Hurt |
| Initialize adjacency from demand-series PCC instead of the proposed initialization (ablation) | CCRNN · NYC Taxi | RMSE: 9.5631 -> 10.0244 | Hurt |
Closest papers by meaning
- Spatiotemporal Multi-Graph Convolution Network for Ride-Hailing Demand Forecasting (AAAI 2019 · similarity 0.90)
- Deep Multi-View Spatial-Temporal Network for Taxi Demand Prediction (AAAI 2018 · similarity 0.89)
- Dynamic Spatial-Temporal Graph Convolutional Neural Networks for Traffic Forecasting (AAAI 2019 · similarity 0.88)
- Community-Aware Multi-Task Transportation Demand Prediction (AAAI 2021 · similarity 0.88)
- TelTrans: Applying Multi-Type Telecom Data to Transportation Evaluation and Prediction via Multifaceted Graph Modeling (AAAI 2024 · similarity 0.84)
- Towards Dynamic Spatial-Temporal Graph Learning: A Decoupled Perspective (AAAI 2024 · similarity 0.84)
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.