Research map / Learning theory and algorithms / Combinatorial and search optimization
Learning to Solve Travelling Salesman Problem with Hardness-Adaptive Curriculum
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 a Gaussian-mixture test distribution with larger c_DIST | Uniform-trained GAT · TSP-50, Gaussian mixture | Optimality Gap (%): 3.5± 0.2 -> 20.0± 1.6 | Hurt |
| Use hardness-adaptive generation instead of uniform sampling | Uniform-pre-trained GAT · TSP-50, generated instances | Optimality Gap (%): 1.70± 0.05 -> 12.36± 1.21 | Helped |
| Increase generator step size η (ablation) | Hardness-adaptive Generator · TSP-50, generated instances | Optimality Gap (%): 0.7± 0.1 -> 1.5± 0.2 -> 2.9± 0.4 -> 12.4± 1.2 -> 14.7± 1.4 | Helped |
| Train with hardness-adaptive samples rather than uniform samples | TSP solver · TSP-50, Gaussian mixture test distribution | Optimality Gap (%) | Helped |
| Add curriculum reweighting based on instance hardness (ablation) | TSP solver trained with hardness-adaptive data · TSP-50, Gaussian mixture test distribution | Optimality Gap (%) | Helped |
Closest papers by meaning
- Unify ML4TSP: Drawing Methodological Principles for TSP and Beyond from Streamlined Design Space of Learning and Search (ICLR 2025 · similarity 0.91)
- Position: Rethinking Post-Hoc Search-Based Neural Approaches for Solving Large-Scale Traveling Salesman Problems (ICML 2024 · similarity 0.88)
- Neural Solver Selection for Combinatorial Optimization (ICML 2025 · similarity 0.87)
- DualOpt: A Dual Divide-and-Optimize Algorithm for the Large-scale Traveling Salesman Problem (AAAI 2025 · similarity 0.87)
- Improving Generalization of Neural Combinatorial Optimization for Vehicle Routing Problems via Test-Time Projection Learning (NeurIPS 2025 · similarity 0.86)
- Destroy and Repair Using Hyper-Graphs for Routing (AAAI 2025 · similarity 0.86)
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.