For SAT and QBF formulas many techniques are applied in order to reduce/modify the number of variables and clauses of the formula, before the formula is passed to the actual solving algorithm. It is well known that these preprocessing techniques often reduce the computation time of the solver by orders of magnitude. In this paper we generalize different preprocessing techniques for SAT and QBF problems to dependency quantified Boolean formulas (DQBF) and describe how they need to be adapted to work with a DQBF solver core. We demonstrate their effectiveness both for CNF- and non-CNF-based DQBF algorithms.
CITATION STYLE
Wimmer, R., Gitina, K., Nist, J., Scholl, C., & Becker, B. (2015). Preprocessing for DQBF. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9340, pp. 173–190). Springer Verlag. https://doi.org/10.1007/978-3-319-24318-4_13
Mendeley helps you to discover research relevant for your work.