Research map / Learning theory and algorithms / Combinatorial and search optimization
Learning to Schedule Heuristics in Branch and Bound
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 learned schedule for diving heuristics instead of default heuristic settings | SCIP 7.0 · GISP, DIMACS graphs | Primal integral: 934.48 -> 470.73 | Helped |
| Use a learned schedule for diving heuristics instead of hand-tuned settings | SCIP 7.0 · GISP, DIMACS graphs | Primal integral: 555.75 -> 470.73 | Helped |
| Use a learned diving-heuristic schedule instead of default settings | SCIP 7.0 · GISP, random graphs | Relative primal integral: 0.76 ± 0.25 -> 0.58 ± 0.20 | Helped |
| Include LNS heuristics in addition to diving heuristics | Learned GISP schedule · GISP, random graphs | Relative primal integral: 0.86 ± 0.28 -> 0.86 ± 0.28 | Helped |
| Replace the proposed greedy schedule search with SMAC | Heuristic scheduling framework · GISP, random graphs | Primal integral | Hurt |
| Train a schedule on smaller instances and apply it to larger instances | Learned diving-heuristic schedule · GISP, random graphs | Relative primal integral: 0.65 ± 0.24 | Helped |
Closest papers by meaning
- L2P-MIP: Learning to Presolve for Mixed Integer Programming (ICLR 2024 · similarity 0.89)
- Parallel Beam Search Algorithms for Domain-Independent Dynamic Programming (AAAI 2024 · similarity 0.85)
- Generalization Guarantees for Learning Score-Based Branch-and-Cut Policies in Integer Programming (NeurIPS 2025 · similarity 0.85)
- SymILO: A Symmetry-Aware Learning Framework for Integer Linear Optimization (NeurIPS 2024 · similarity 0.84)
- Differentiable Combinatorial Scheduling at Scale (ICML 2024 · similarity 0.84)
- MILP-StuDio: MILP Instance Generation via Block Structure Decomposition (NeurIPS 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.