Efficient Encoding of Cost Optimal Delete-Free Planning as SAT
Authors: Masood Feyzbakhsh Rankooh, Jussi Rintanen9910-9917
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our empirical results indicate that this method is quite competitive with the state of the art, demonstrating a better coverage compared to that of competing methods on standard STRIPS planning benchmark problems. Figures 1 and 2 show the comparison of the time used for solving benchmark instances for the mentioned three versions. |
| Researcher Affiliation | Academia | Department of Computer Science, Aalto University, Helsinki, Finland |
| Pseudocode | No | The paper does not contain any sections or figures explicitly labeled 'Pseudocode' or 'Algorithm', nor does it present structured steps formatted like code. |
| Open Source Code | No | The paper does not provide an unambiguous statement or a direct link indicating that the source code for the methodology described in this paper is publicly available. |
| Open Datasets | Yes | As our benchmark problem set, we have used the deletefree versions of all STRIPS planning problem sets found in the planning repository1. From IPC domains, domains from both satisficing and optimal tracks have been considered. 1https://github.com/AI-Planning/classical-domains |
| Dataset Splits | No | The paper mentions using a 'benchmark problem set' but does not provide specific details about training, validation, or test dataset splits (e.g., percentages, sample counts, or references to predefined splits). |
| Hardware Specification | No | The paper states: 'All experiments have been run on a cluster of Linux machines,' which is a general statement. It does not provide specific details about CPU models, GPU models, memory, or other hardware components used. |
| Software Dependencies | Yes | In all cases we have used Kissat (Biere et al. 2020) as the SAT solver. ... As the optimizer for these methods we have used IBM ILOG CPLEX Optimization Studio 20.1. |
| Experiment Setup | Yes | For producing the optimal cost for the given problem, we perform a search mechanism similar to binary search method. First, for u = 1, 2, 4, 8, .. formulas with upper bound u on the total cost are produced and given sequentially to the SAT solver until a satisfiable formula is found. ... All of our implemented versions use some of the preprocessing methods presented in (Imai and Fukunaga 2015). The used preprocessing methods are 1) finding fact landmarks (Gefen and Brafman 2011) and adding them to the goal conditions; 2) doing action relevance analysis and removing non-relevant actions; and 3) dominated action elimination. ... for determining the order of vertex elimination, we have implemented the minimum degree heuristic, i.e., eliminating a vertex with minimal total number of incoming and outgoing edges in the graph produced after the elimination of previously eliminated vertices. |