The separation of planner logic from domain knowledge supports the use of reformulation and configuration techniques, such as macro-actions and entanglements, which transform the model representation in order to improve a planner’s performance. One drawback of such an approach is that it may require a potentially expensive training phase. In this paper, we introduce heuristic approaches for the online configuration of planning domain models. The proposed heuristics consider different aspects of PDDL-encoded operators for reordering such operators in the domain model, relying on the assumption that the way in which operators are encoded carries useful information about their expected use.
|Title of host publication||Proceedings of the Tenth International Symposium on Combinatorial Search (SoCS 2017)|
|Editors||Alex Fukunaga, Akihiro Kishimoto|
|Number of pages||2|
|ISBN (Print)||9781577357902, 1577357906|
|Publication status||Published - 15 Jun 2017|
|Event||10th Annual Symposium on Combinatorial Search - Pittsburgh, United States|
Duration: 16 Jun 2017 → 17 Jun 2017
Conference number: 10
http://socs17.dreamhosters.com/ (Link to Symposium Website )
|Conference||10th Annual Symposium on Combinatorial Search|
|Abbreviated title||SoCS 2017|
|Period||16/06/17 → 17/06/17|
Vallati, M., Chrpa, L., & McCluskey, T. (2017). Improving a planner’s performance through online heuristic configuration of domain models. In A. Fukunaga, & A. Kishimoto (Eds.), Proceedings of the Tenth International Symposium on Combinatorial Search (SoCS 2017) AAAI press.