Automated Verification of Social Laws in Numeric Settings
Authors: Ronen Nir, Alexander Shleyfman, Erez Karpas
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We empirically verify robustness in multiple domains using state-of-the-art numeric planners. Additionally, this compilation raises a challenge by generating a set of non-trivial numeric domains where unsolvability should be either proved or disproved. |
| Researcher Affiliation | Academia | Ronen Nir1, Alexander Shleyfman2, Erez Karpas1 1 Technion Israel Institute of Technology 2 Bar-Ilan University |
| Pseudocode | No | The paper describes methods and actions using formal notation and text but does not include any clearly labeled 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | Yes | We implemented the compilation in Python2 and tested it on domains from previous work on numeric planning along with the BRIDGE domain that we have formulated to illustrate an interesting view of how numerical SLs are more compact than SLs in classical planning settings. (...) 2https://github.com/ronen85/numeric-slv |
| Open Datasets | Yes | The domains DEPOTS, SAILING, PLANT-WATERING (Scala et al. 2016) and DELIVERY (Shleyfman et al. 2022) were used to demonstrate the scalability of the compilation. |
| Dataset Splits | No | The paper mentions using several domains for evaluation but does not provide specific details on training, validation, or test dataset splits (e.g., percentages, sample counts, or predefined split references). |
| Hardware Specification | Yes | We used a single Intel i7-7700K core on a computer with 32GB of RAM. |
| Software Dependencies | Yes | We implemented the compilation in Python2 and tested it on domains from previous work on numeric planning along with the BRIDGE domain that we have formulated to illustrate an interesting view of how numerical SLs are more compact than SLs in classical planning settings. Our compilation takes a numeric multi-agent problem defined in PDDL2.1 (Fox and Long 2003) and an additional JSON file with information about the agents goal affiliation and waitfor conditions. A numerical planning problem is then generated. To solve the generated problems, we used the metricfast forward planner (Hoffmann 2003b). Its performance has been demonstrated to be better than that of other planners, such as NFD (Aldinger and Nebel 2017) and ENHSP20 (Scala 2020). |
| Experiment Setup | Yes | The time limit is 30 minutes. |