Generalized Planning via Abstraction: Arbitrary Numbers of Objects
Authors: León Illanes, Sheila A. McIlraith7610-7618
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We report experimental results that show our approach is a practical and promising technique for solving an interesting class of problems. We evaluate LOOM over a suite of generalized planning problems. The results show our approach is capable of quickly finding general solutions, and that these can be instantiated into plans for specific problems orders of magnitude faster than planning from scratch, without significant changes in plan quality. |
| Researcher Affiliation | Academia | Le on Illanes, Sheila A. Mc Ilraith Department of Computer Science University of Toronto, Toronto, Canada {lillanes, sheila}@cs.toronto.edu |
| Pseudocode | Yes | Algorithm 1: The LOOM planning algorithm. Algorithm 2: The termination verification procedure used by LOOM. |
| Open Source Code | No | The paper states 'We implemented our main LOOM algorithm by modifying the existing codebase for FOND planner PRP1'. The provided URL (https://bitbucket.org/haz/planner-for-relevant-policies) is for the PRP planner, not explicitly for the authors' modifications or the LOOM implementation itself. |
| Open Datasets | No | The paper mentions evaluating LOOM on domains such as 'Recycling', 'Logistics', 'Hamburger', 'Construction', and 'Roundabout'. While these are referred to as from 'existing literature' or 'well-known classical planning problems', the paper does not provide specific links, DOIs, repositories, or formal citations (with author names and year) for these datasets to confirm their public availability or how to access them. |
| Dataset Splits | No | The paper describes experiments but does not provide specific details on training, validation, or test dataset splits, percentages, or sample counts. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used to run the experiments, such as CPU or GPU models, or memory specifications. |
| Software Dependencies | No | The paper mentions modifying the codebase for 'FOND planner PRP' and using 'Fast Downward Planning System' and 'LAMA heuristics', but it does not provide specific version numbers for these software components or any other ancillary software. |
| Experiment Setup | No | The paper does not provide specific experimental setup details such as hyperparameter values, training configurations, or system-level settings for the algorithms used. |