Research map / Learning theory and algorithms / Combinatorial and search optimization
Narrowing the Gap Between Saturated and Optimal Cost Partitioning for Classical Planning
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 |
|---|---|---|---|
| optimize the abstraction order by hill climbing instead of a random order | SCP · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 759.3 -> 884.9 | Helped |
| hill-climbed order instead of the h-add orders | SCP · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 800 -> 884.9 | Helped |
| maximize over 200 random orders instead of a single random order (ablation) | SCP · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 759.3 -> 947.0 | Helped |
| increase random orders beyond 200 (to 1000) (ablation) | SCP max over random orders · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 947.0 -> 879.5 | Hurt |
| use multiple hill-climbed orders instead of random orders (ablation) | SCP max over orders · IPC 1998-2014 optimal planning (1667 tasks) | coverage | No clear effect |
| keep only orders that add information on sample states (diversification) instead of 200 random orders (ablation) | SCP · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 947.0 -> 966.7 | Helped |
| increase diversification time from 0.1s to 200s (ablation) | SCP div · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 928.3 -> 966.7 | Helped |
| increase diversification time beyond 200s to 1200s (ablation) | SCP div · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 966.7 -> 956.3 | Hurt |
| SCP div instead of the strongest other admissible heuristic (hSEQ+LM-cut) | optimal planning heuristics · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 895 -> 966.7 | Helped |
| SCP div instead of optimal cost partitioning under 30 min / 2GiB vs 24h / 3.5GiB | cost partitioning · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 531 -> 966.7 | Helped |
| SCP div with h2 mutex pruning instead of SymBA*2 | optimal planning · IPC 1998-2014 optimal planning (1667 tasks) | coverage: 1008 -> 1017.4 | No clear effect |
Closest papers by meaning
- Saturated Post-hoc Optimization for Classical Planning (AAAI 2021 · similarity 0.91)
- Not Everything Is Permitted: Constrained Cartesian Abstractions for Optimal Classical Planning (AAAI 2026 · similarity 0.89)
- A* Search and Bound-Sensitive Heuristics for Oversubscription Planning (AAAI 2022 · similarity 0.87)
- When CEGAR Meets Regression: A Love Story in Optimal Classical Planning (AAAI 2024 · similarity 0.87)
- Novelty vs. Potential Heuristics: A Comparison of Hardness Measures for Satisficing Planning (AAAI 2024 · similarity 0.86)
- PDBs Go Numeric: Pattern-Database Heuristics for Simple Numeric Planning (AAAI 2025 · 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.