The Fewer the Merrier: Pruning Preferred Operators with Novelty
Authors: Alexander Tuisov, Michael Katz
IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experimental evaluation shows the practical benefit of our suggested approach, compared to the currently used methods. |
| Researcher Affiliation | Collaboration | 1Technion, Haifa, Israel 2IBM Research, Yorktown Heights, NY, USA |
| Pseudocode | No | No structured pseudocode or algorithm blocks were found in the paper. The methods are described in narrative text and formal definitions. |
| Open Source Code | Yes | The code is at https://github.com/IBM/FD-Novelty-PO |
| Open Datasets | Yes | The benchmark set consists of all STRIPS benchmarks from the satisficing tracks of International Planning Competitions (IPC) 1998-2018, a total of 1816 tasks in 64 domains. |
| Dataset Splits | No | The paper mentions using 'all STRIPS benchmarks from the satisficing tracks of International Planning Competitions (IPC) 1998-2018', but does not specify training, validation, or test dataset splits in the typical machine learning sense. |
| Hardware Specification | Yes | The experiments were performed on Intel(R) Xeon(R) Gold 6248 CPU @2.50GHz machines, with the time and memory limit of 30min and 4GB, respectively |
| Software Dependencies | No | The paper states, 'we implemented it on top of the Fast Downward planning system [Helmert, 2006].' However, it does not provide specific version numbers for Fast Downward or any other software dependencies. |
| Experiment Setup | Yes | All tested configuration perform a greedy best-first search with delayed evaluation and multiple queues. ... with the time and memory limit of 30min and 4GB, respectively. ... We select top k elements and test three bounds, k {10, 100, 1000}. |