Research map / Probabilistic and scientific modeling / Molecules and proteins
Simulation-Driven Railway Delay Prediction: An Imitation Learning Approach
Clément Elliker, Jesse Read, Sonia Vanier, Albert Bifet
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 |
|---|---|---|---|
| train by simulation (BC) instead of regression | Transformer · INFRABEL delays | MAE: 60.53 -> 56.17 | Improved |
| add drift-corrected imitation learning (DCIL) over BC (ablation) | Transformer · INFRABEL delays | MAE: 56.17 -> 52.24 | Improved |
| add DCIL over BC (ablation) | MLP · INFRABEL delays | MAE: 59.10 -> 57.76 | Improved |
| train by simulation (BC) instead of regression | XGBoost · INFRABEL delays | MAE: 64.48 -> 66.77 | Hurt |
| use MLP with imitation learning instead of Transformer with regression | MLP (1.4M params) vs Transformer (19M) · INFRABEL delays | MAE: 60.53 -> 57.76 | Improved |
| add DCIL over BC at long horizon (ablation) | Transformer · INFRABEL 25-30 min | MAE: 87.75 -> 83.38 | Improved |
| replace BC/DCIL with GAIL adversarial imitation | Transformer · INFRABEL delays | MAE | Hurt |
| use BC with XGBoost at short vs long horizons | XGBoost · INFRABEL horizons | MAE: 31.31 -> 27.95 (0-5 min); 95.94 -> 105.31 (25-30 min) | Mixed |
Closest papers by meaning
- LASIL: Learner-Aware Supervised Imitation Learning For Long-term Microscopic Traffic Simulation (CVPR 2024 · similarity 0.84)
- Responsive Dynamic Graph Disentanglement for Metro Flow Forecasting (AAAI 2025 · similarity 0.82)
- Stable Long-Horizon PDE Forecasting via Latent Structured Spectral Propagators (NeurIPS 2026 · similarity 0.82)
- Active Sequential Posterior Estimation for Sample-Efficient Simulation-Based Inference (NeurIPS 2024 · similarity 0.82)
- Improving Transferability for Cross-Domain Trajectory Prediction via Neural Stochastic Differential Equation (AAAI 2024 · similarity 0.82)
- Directly Forecasting Belief for Reinforcement Learning with Delays (ICML 2025 · similarity 0.82)
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.
