Research map / Learning theory and algorithms / Combinatorial and search optimization
A Box-Constrained Approach for Hard Permutation Problems
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 |
|---|---|---|---|
| raise FAQ Frank-Wolfe iteration cap from 10 to 30 | FAQ · QAPLIB CHR | best-run optimality gap: 18.72 -> 9.32 | Helped |
| add greedy swap heuristic to FAQ30 | FAQ30 · QAPLIB CHR | best-run optimality gap: 9.32 -> 5.90 | Helped |
| use CD+C+RS instead of FAQ30 | FAQ30 · QAPLIB ELS | best-run optimality gap: 22.96 -> 4.21 | Helped |
| use CD+C+RS instead of FAQ10 | FAQ10 · QAPLIB LIPA B | best-run optimality gap: .00 -> 7.92 | Hurt |
| use CD+C+RS instead of FAQ30 | FAQ30 · QAPLIB TAI | best-run optimality gap: 1.53 -> 1.30 | Mixed |
| use CD+C+RS instead of FAQ30+G | FAQ30+G · QAPLIB NUG | best-run optimality gap: .03 -> .43 | Hurt |
| use CD+C+RS instead of FAQ10 (running time) | FAQ10 · QAPLIB ELS | timing ratio: 394.71 | Helped |
| use CD+C+RS instead of FAQ30+G (running time, synthetic QAP) | FAQ30+G · synthetic Taillard QAP n=300..1000 | running time: about 60 times faster | Helped |
| use CD+C+RS instead of FAQ10+G (running time, synthetic QAP) | FAQ10+G · synthetic Taillard QAP n=300..1000 | running time: about 20 times faster | Helped |
| CD+C+RS vs FAQ30+G objective on synthetic QAP | FAQ30+G · synthetic Taillard QAP | objective gap | Hurt |
| replace random sorting network by full sorting network in heuristic (ablation) | CD+C · QAPLIB | objective value | Helped |
| replace random sorting network by full sorting network (ablation) | CD+C · synthetic QAP | run time scaling: O(n^2) -> O(n^3) | Hurt |
Closest papers by meaning
- Differentiable extensions with rounding guarantees for combinatorial optimization over permutations (NeurIPS 2025 · similarity 0.88)
- Polynomial Time Learning Augmented Algorithms for NP-hard Permutation Problems (ICML 2025 · similarity 0.86)
- OT4P: Unlocking Effective Orthogonal Group Path for Permutation Relaxation (NeurIPS 2024 · similarity 0.85)
- Graph Alignment via Birkhoff Relaxation (NeurIPS 2025 · similarity 0.85)
- Differentiable Quadratic Optimization For the Maximum Independent Set Problem (ICML 2025 · similarity 0.84)
- Learning Solution-Aware Transformers for Efficiently Solving Quadratic Assignment Problem (ICML 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.