Generalized Label Reduction for Merge-and-Shrink Heuristics
Abstract
Label reduction is a technique for simplifying families of labeled transition systems by dropping distinctions between certain transition labels. While label reduction is critical to the efficient computation of merge-and-shrink heuristics, current theory only permits reducing labels in a limited number of cases. We generalize this theory so that labels can be reduced in every intermediate abstraction of a merge-and-shrink tree. This is particularly important for efficiently computing merge-and-shrink abstractions based on non-linear merge strategies. As a case study, we implement a non-linear merge strategy based on the original work on merge-and-shrink heuristics in model checking by Dräger et al.
Cite
Text
Sievers et al. "Generalized Label Reduction for Merge-and-Shrink Heuristics." AAAI Conference on Artificial Intelligence, 2014. doi:10.1609/AAAI.V28I1.9028Markdown
[Sievers et al. "Generalized Label Reduction for Merge-and-Shrink Heuristics." AAAI Conference on Artificial Intelligence, 2014.](https://mlanthology.org/aaai/2014/sievers2014aaai-generalized/) doi:10.1609/AAAI.V28I1.9028BibTeX
@inproceedings{sievers2014aaai-generalized,
title = {{Generalized Label Reduction for Merge-and-Shrink Heuristics}},
author = {Sievers, Silvan and Wehrle, Martin and Helmert, Malte},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2014},
pages = {2358-2366},
doi = {10.1609/AAAI.V28I1.9028},
url = {https://mlanthology.org/aaai/2014/sievers2014aaai-generalized/}
}