The Complexity of LTL on Finite Traces: Hard and Easy Fragments

Authors: Valeria Fionda, Gianluigi Greco

AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Motivated by the theoretical results we implemented a reasoner for LTLf, called LTL2SAT1, and compared it with Aalta (Li et al. 2014)... We used the same datasets used in (Li et al. 2014)... Experiments have been executed on a PC Intel Core i5 2,4 GHz, 8GB RAM. For each formula, we measured the ratio between the time required by LTL2SAT to check satisfiability and that required by Aalta. Figure 4 reports the results as percentage stacked bar charts, for each dataset.
Researcher Affiliation Academia Valeria Fionda and Gianluigi Greco De Ma CS, University of Calabria, Italy {fionda, ggreco}@mat.unical.it
Pseudocode No The paper describes formal frameworks and an encoding approach but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes Motivated by the theoretical results we implemented a reasoner for LTLf, called LTL2SAT1... Downloadable at http://ltl2sat.wordpress.com/
Open Datasets Yes We used the same datasets used in (Li et al. 2014)
Dataset Splits No The paper mentions using datasets for comparison but does not provide specific details on training, validation, or test splits such as percentages or sample counts.
Hardware Specification Yes Experiments have been executed on a PC Intel Core i5 2,4 GHz, 8GB RAM.
Software Dependencies No LTL2SAT rewrites the input formula into an equivalent Boolean formula and uses glucose (Audemard and Simon 2009) to compute a model. The specific version number for glucose is not provided.
Experiment Setup Yes When the optimization strategies are not applicable, the size n of the trace to be used in the SAT rewriting is initially set to 1: If no model is found, then n is doubled and the process is repeated until n exceeds the theoretical upper bound (or a 20s time-limit is reached).