Research map / Reinforcement learning / Visual and representation RL
Ensemble-based Deep Reinforcement Learning for Vehicle Routing Problems under Distribution Shift
Yuan Jiang, Zhiguang Cao, Yaoxin Wu, Wen Song, Jie Zhang
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 |
|---|---|---|---|
| remove bootstrap with random initialization (BRI) (ablation) | EL-DRL · CVRP100 | avg gap: 2.201% -> 2.055% | Improved |
| add Theil index regularization (ablation) | EL-DRL · CVRP100 | avg gap: 2.090% -> 2.055% | Improved |
| add parameter dissimilarity regularization (ablation) | EL-DRL · CVRP100 | avg gap: 2.118% -> 2.055% | Improved |
| stack four POMO copies without diversity mechanisms (Multi-POMO) (ablation) | POMO · CVRP100 | avg gap: 2.423% -> 2.311% | Improved |
| use EL-DRL ensemble instead of single POMO (ablation) | POMO · CVRP100 | avg gap: 2.423% -> 2.055% | Improved |
| EL-DRL instead of AM | AM · TSP100 Cluster | gap: 4.41% -> 2.07% | Improved |
| EL-DRL instead of POMO | POMO · TSP100 Cluster | gap: 2.62% -> 2.07% | Improved |
| multiple decoders with KL divergence (MDAM) instead of AM | AM · TSP100 | gap: 4.41% -> 3.72% | No clear effect |
| EL-DRL instead of AMDKD | AMDKD · TSPLib | gap: 5.30% -> 4.85% | Improved |
| EL-DRL instead of AMDKD | AMDKD · CVRPLib | gap: 4.32% -> 4.12% | Improved |
| EL-DRL instead of POMO | POMO · CVRP50 Cluster | gap: 1.35% -> 0.87% | Improved |
| EL-DRL instead of HAC | HAC · TSP100 | gap: 5.10% -> 2.07% | Improved |
Closest papers by meaning
- MVMoE: Multi-Task Vehicle Routing Solver with Mixture-of-Experts (ICML 2024 · similarity 0.86)
- Memory-Enhanced Neural Solvers for Routing Problems (NeurIPS 2025 · similarity 0.86)
- MTL-KD: Multi-Task Learning Via Knowledge Distillation for Generalizable Neural Vehicle Routing Solver (NeurIPS 2025 · similarity 0.86)
- One Step is Enough: Multi-Agent Reinforcement Learning Based on One-Step Policy Optimization for Order Dispatch on Ride-Sharing Platforms (NeurIPS 2026 · similarity 0.86)
- Improving Generalization of Neural Combinatorial Optimization for Vehicle Routing Problems via Test-Time Projection Learning (NeurIPS 2025 · similarity 0.85)
- Collaboration! Towards Robust Neural Methods for Routing Problems (NeurIPS 2024 · similarity 0.85)
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.
