Communication Dans Un Congrès Année : 2025

Learning Efficiency Meets Symmetry Breaking

Résumé

Learning-based planners leveraging Graph Neural Networks can learn search guidance applicable to large search spaces, yet their potential to address symmetries remains largely unexplored. In this paper, we introduce a graph representation of planning problems allying learning efficiency with the ability to detect symmetries, along with two pruning methods, action pruning and state pruning, designed to manage symmetries during search. The integration of these techniques into Fast Downward achieves a first-time success over LAMA on the latest IPC learning track dataset.

Fichier principal
Vignette du fichier
icaps25.pdf (278.41 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence
DOI

Cite 10.5281/zenodo.6382173 Article Seipp, J., Torralba, Á., & Hoffmann, J. (2022). PDDL Generators. Zenodo. https://doi.org/10.5281/ZENODO.6382173

Planification – Générateurs PDDL (données d’ablation)

Dates et versions

hal-05226738 , version 1 (27-08-2025)

Licence

Identifiants

  • HAL Id : hal-05226738 , version 1

Citer

Yingbin Bai, Sylvie Thiébaux, Felipe Trevizan. Learning Efficiency Meets Symmetry Breaking. International Conference on Automated Planning and Scheduling (ICAPS-25), Nov 2025, Melbourne, Australia. ⟨hal-05226738⟩
627 Consultations
267 Téléchargements

Partager

  • More