Planning with Qualitative Action-Trajectory Constraints in PDDL

Authors: Luigi Bonassi, Alfonso Emilio Gerevini, Enrico Scala

IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we experimentally study the usefulness of our action-trajectory constraints as a tool to express control knowledge. The experimental results show that the performance of a classical planner can be significantly improved by exploiting knowledge expressed by action constraints and handled by our compilation
Researcher Affiliation Academia Luigi Bonassi , Alfonso Emilio Gerevini and Enrico Scala Dipartimento di Ingegneria dell Informazione, Universit a degli Studi di Brescia, Italy
Pseudocode Yes Algorithm 1: PAC-C
Open Source Code Yes Full proof in the supplementary material.3 3Supplementary material, benchmark domains and Python implementation of PAC-C can be found at https://bit.ly/3kerz8s.
Open Datasets Yes Since there are no available benchmarks featuring action constraints, we generated a new benchmark suite3 starting from the problems of the 5th International Planning Competition. ... 3Supplementary material, benchmark domains and Python implementation of PAC-C can be found at https://bit.ly/3kerz8s.
Dataset Splits No The paper describes using a benchmark suite for evaluation and does not specify training, validation, or test dataset splits in the conventional machine learning sense.
Hardware Specification Yes All experiments ran on an Xeon Gold 6140M 2.3 GHz, with time and memory limits of 1800s and 8GB, respectively.
Software Dependencies No The paper mentions software like LAMA, PACC3, TCORE, and LTL-C, but it does not provide specific version numbers for these software components.
Experiment Setup Yes All experiments ran on an Xeon Gold 6140M 2.3 GHz, with time and memory limits of 1800s and 8GB, respectively.