edoc

State-dependent Cost Partitionings for Cartesian Abstractions in Classical Planning

Keller, Thomas and Pommerening, Florian and Seipp, Jendrik and Geißer, Florian and Mattmüller, Robert. (2016) State-dependent Cost Partitionings for Cartesian Abstractions in Classical Planning. In: Proceedings of the 25th International Joint Conference on Artificial Intelligence (IJCAI 2016), 4. Palo Alto, pp. 3161-3169.

[img] PDF - Published Version
928Kb

Official URL: http://edoc.unibas.ch/45218/

Downloads: Statistics Overview

Abstract

Abstraction heuristics are a popular method to guide optimal search algorithms in classical planning. Cost partitionings allow to sum heuristic estimates admissibly by distributing action costs among the heuristics. We introduce state-dependent cost partitionings which take context information of actions into account, and show that an optimal state-dependent cost partitioning dominates its state-independent counterpart. We demonstrate the potential of our idea with a state-dependent variant of the recently proposed saturated cost partitioning, and show that it has the potential to improve not only over its state-independent counterpart, but even over the optimal state-independent cost partitioning. Our empirical results give evidence that ignoring the context of actions in the computation of a cost partitioning leads to a significant loss of information.
Faculties and Departments:05 Faculty of Science > Departement Mathematik und Informatik > Informatik > Artificial Intelligence (Helmert)
UniBasel Contributors:Keller, Thomas and Pommerening, Florian and Seipp, Jendrik
Item Type:Conference or Workshop Item, refereed
Conference or workshop item Subtype:Conference Paper
Publisher:IJCAI/AAAI Press
ISBN:978-1-57735-777-3
Note:Publication type according to Uni Basel Research Database: Conference paper
Language:English
Related URLs:
edoc DOI:
Last Modified:05 Apr 2017 12:17
Deposited On:13 Dec 2016 13:50

Repository Staff Only: item control page