LTLf Synthesis as AND-OR Graph Search: Knowledge Compilation at Work

Authors: Giuseppe De Giacomo, Marco Favorito, Jianwen Li, Moshe Y. Vardi, Shengping Xiao, Shufang Zhu

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

Reproducibility Variable Result LLM Response
Research Type Experimental We implemented our approach in a tool called Cynthia and conducted comprehensive experiments by comparing to existing LTLf synthesis tools, including Lisa, Lydia and Ltlfsyn from [Xiao et al., 2021] and demonstrate the merits of our approach.
Researcher Affiliation Collaboration 1Sapienza University of Rome 2Banca d Italia 3East China Normal University 4Rice University
Pseudocode Yes The paper includes clearly labeled algorithm blocks: 'Algorithm 1 SDD-based Forward Synthesis', 'Algorithm 2 Propagate Success Backwards', and 'Algorithm 3 SDD-based Expand Graph from An Or Node'.
Open Source Code Yes 1Tool available at https://whitemech.github.io/cynthia.
Open Datasets Yes We collected, in total, 1494 LTLf synthesis instances from literature: 40 Patterns instances [Xiao et al., 2021]; 54 Two-player-Games instances [Tabajara and Vardi, 2019; Bansal et al., 2020]; 1400 Random instances [Zhu et al., 2017; De Giacomo and Favorito, 2021].
Dataset Splits No The paper mentions using LTLf synthesis instances as benchmarks for evaluation, but it does not specify any training, validation, or test dataset splits (e.g., percentages or sample counts) for reproducibility.
Hardware Specification Yes Experiments were run on a computer cluster, where each instance took exclusive access to a computing node with Intel-Xeon processor running at 2.6 GHz, with 8GB of memory and 300 seconds of time limit.
Software Dependencies Yes We make use of library SDD-2.0 (http://reasoning.cs.ucla.edu/sdd) to handle all SDD related operations.
Experiment Setup No The paper describes the algorithms and their implementation details, and mentions some optimizations ('pre-processing techniques', 'look-ahead check'), but it does not provide specific hyperparameter values, training configurations, or other system-level settings for the experiments.