Acting and Planning Using Operational Models

Authors: Sunandita Patra, Malik Ghallab, Dana Nau, Paolo Traverso7691-7698

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experiments show significant benefits in the efficiency of the acting and planning system. We have implemented and tested our framework on four domains.
Researcher Affiliation Academia 1Department of Computer Science and Institute for Systems Research, University of Maryland, College Park, USA 2Centre national de la recherche scientifique (CNRS), Toulouse, France 3Fondazione Bruno Kessler (FBK), Povo Trento, Italy
Pseudocode Yes Algorithm 1: RAE (Refinement Acting Engine)
Open Source Code Yes Full code is online at https://bitbucket.org/sunandita/raeplan.
Open Datasets No The paper describes experimental domains (Explorable Environment, Chargeable Robot, Spring Door, Industrial Plant) and test suites but does not provide specific links, DOIs, or citations to publicly available datasets used within these domains.
Dataset Splits No The paper discusses running problems multiple times but does not specify train, validation, or test dataset splits in terms of percentages, absolute counts, or predefined splits.
Hardware Specification Yes running on a 2.6 GHz Intel Core i5 processor.
Software Dependencies No The paper does not mention specific software dependencies with version numbers (e.g., Python, PyTorch, or specific solvers/libraries).
Experiment Setup Yes We ran experiments with k = 0, 3, 5, 7, 10. In the CR, EE and IP domains we used b = 1, 2, 3 because each task are at most three method instances. In the SD domain, we used b = 1, 2, 3, 4 because it has four methods for opening a door.