Research map / Graph learning / GNN architectures
Online Planner Selection with Graph Neural Networks and Adaptive Scheduling
Tengfei Ma, Patrick Ferber, Siyu Huo, Jie Chen, Michael Katz
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 |
|---|---|---|---|
| Use lifted rather than grounded task graphs with GCN (ablation) | GCN planner selection · IPC 2018 test set, Delfi split | solved tasks: 80.7% -> 87.6% | Improved |
| Use lifted rather than grounded task graphs with GG-NN (ablation) | GG-NN planner selection · IPC 2018 test set, Delfi split | solved tasks: 77.9% -> 81.4% | Improved |
| Replace CNN with GCN on lifted graphs (ablation) | Online planner selection · IPC 2018 test set, Delfi split | solved tasks: 86.9% -> 87.6% | Improved |
| Use GCN rather than GG-NN on lifted graphs (ablation) | Online planner selection · IPC 2018 test set, domain-preserving splits | solved tasks: 76.6% -> 85.6% | Improved |
| Use GCN rather than GG-NN on lifted graphs (ablation) | Online planner selection · IPC 2018 test set, random splits | solved tasks: 74.4% -> 87.2% | Improved |
| Add halftime adaptive planner switching (ablation) | GCN planner selection · IPC 2018 test set, Delfi split | solved tasks: 87.6% -> 89.7% | Improved |
| Add halftime adaptive planner switching (ablation) | GCN planner selection · IPC 2018 test set, domain-preserving splits | solved tasks: 85.6% -> 91.1% | Improved |
| Add halftime adaptive planner switching (ablation) | GCN planner selection · IPC 2018 test set, random splits | solved tasks: 87.2% -> 92.1% | Improved |
| Add halftime adaptive planner switching (ablation) | GG-NN planner selection · IPC 2018 test set, domain-preserving splits | solved tasks: 76.6% -> 83.0% | Improved |
| Add halftime adaptive planner switching (ablation) | GG-NN planner selection · IPC 2018 test set, random splits | solved tasks: 74.4% -> 86.6% | Improved |
| Increase the number of fixed-schedule planners from 2 to 3 (ablation) | Greedy offline planner scheduling · IPC 2018 test set, Delfi split | solved tasks: 85.5% -> 92.4% | Improved |
| Increase the number of fixed-schedule planners from 3 to 4 (ablation) | Greedy offline planner scheduling · IPC 2018 test set, Delfi split | solved tasks: 92.4% -> 89.7% | Hurt |
Closest papers by meaning
- State Encodings for GNN-Based Lifted Planners (AAAI 2025 · similarity 0.88)
- Graph Neural Network Based Action Ranking for Planning (NeurIPS 2025 · similarity 0.87)
- Graph Learning for Numeric Planning (NeurIPS 2024 · similarity 0.86)
- Learning Domain-Independent Heuristics for Grounded and Lifted Planning (AAAI 2024 · similarity 0.86)
- Learning More Expressive General Policies for Classical Planning Domains (AAAI 2025 · similarity 0.84)
- Can Graph Learning Improve Planning in LLM-based Agents? (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.
