Refining HTN Methods via Task Insertion with Preferences
Authors: Zhanhao Xiao, Hai Wan, Hankui Hankz Zhuo, Andreas Herzig, Laurent Perrussel, Peilin Chen10009-10016
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental Analysis In this section, we evaluate METHODREFINE1 in three wellknown planning domains comparing with HTN-MAKER2 on the ability of solving new instances. |
| Researcher Affiliation | Academia | Zhanhao Xiao,1 Hai Wan,1* Hankui Hankz Zhuo,1 Andreas Herzig,2,3 Laurent Perrussel,3 Peilin Chen1 1School of Data and Computer Science, Sun Yat-sen University, Guangzhou, China 2IRIT, CNRS, Toulouse, France 3University of Toulouse, Toulouse, France |
| Pseudocode | Yes | Algorithm 1: COMPLETE(σ, T , P) |
| Open Source Code | Yes | Available in https://github.com/sysulic/MethodRefine |
| Open Datasets | Yes | We first get the problem generators from International Planning Competition website3 and randomly generate 100 instances for each domain and take 50 instances as the training set and 50 instances as the testing set. |
| Dataset Splits | No | The paper states "take 50 instances as the training set and 50 instances as the testing set." but does not mention a validation set or details on how validation was performed. |
| Hardware Specification | Yes | Experiments are run on the 3.00 GHz Intel i5-7400 with 8 GB RAM with a cutoff time of one hour. |
| Software Dependencies | No | The paper mentions 'HTN-MAKER' and 'SHOP2 HTN planner' but does not provide specific version numbers for any software dependencies. |
| Experiment Setup | Yes | We run METHODREFINE and HTN-MAKER with 50 instances growingly as input and obtain different learned method sets from these two approaches. ... To simulate the incomplete method set as the input of METHODREFINE, we take the HTN domain descriptions in the website4 of SHOP2 HTN planner, and remove different sets of subtasks from these domains. Then we consider three removal cases: 1) remove one primitive task from each method (if exists), with meaning the high completeness, noted by MR-H; 2) remove two primitive tasks from each method (if exists), noted by MR-M, with meaning the middle completeness; 3) remove one more compound task in some method of MR-L, noted by MR-L, with meaning the low completeness. |