Solving Goal Recognition Design Using ASP

Authors: Tran Son, Orkunt Sabuncu, Christian Schulz-Hanke, Torsten Schaub, William Yeoh

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experimental results show that one of our ASP encodings is more scalable and is significantly faster by up to three orders of magnitude than the current state of the art.
Researcher Affiliation Academia New Mexico State University, Las Cruces, NM, USA University of Potsdam, Germany INRIA, Rennes, France
Pseudocode Yes Algorithm 1 GRD( D, G , k, max)
Open Source Code No The paper describes its proposed ASP encodings and algorithms but does not provide a direct link to the source code or explicitly state that the code is publicly available.
Open Datasets Yes We also used the same four benchmark domains that they have made publicly available:1 (1) GRID-NAVIGATION, where each instance is defined by the xand y-dimensions; (2) IPC-GRID+, where each instance is defined by the xand y-dimensions and the number of locks/keys; (3) BLOCKWORDS, where each instance is defined by the number of blocks and words/goals; and (4) LOGISTICS, where each instance is defined by the number of airplanes, airports, locations, cities, trucks, and packages. 1http://technion.ac.il/ sarahn/final-benchmarks-icaps-2014/.
Dataset Splits No The paper mentions using benchmark domains for evaluation but does not specify explicit dataset splits for training, validation, or testing, nor does it describe any cross-validation setup.
Hardware Specification Yes We set k={1, 2} as suggested in the benchmarks, conducted our experiments on a 3.60GHz CPU machine with 8GB of RAM, and set a timeout of 5 hours.
Software Dependencies No The paper mentions software like 'ASP', 'clingo', 'Python procedures', and 'classical planner' but does not provide specific version numbers for these dependencies, which are necessary for reproducibility.
Experiment Setup Yes We set k={1, 2} as suggested in the benchmarks, conducted our experiments on a 3.60GHz CPU machine with 8GB of RAM, and set a timeout of 5 hours. GRD(.) controls the computation of wcd(P) and a solution of P wrt a given k assuming that the maximal length of plans to all goals in G is at most max by (i) setting the bound of plan cost (max, Line 3 6), (ii) setting the parameter len of π(P) (Line 8).