G2SAT: Learning to Generate SAT Formulas

G2SAT: Learning to Generate SAT Formulas” by Jiaxuan You, Haoze Wu, Clark Barrett, Raghuram Ramanujan, and Jure Leskovec. In Advances in Neural Information Processing Systems 32 (NeurIPS '19), (H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché-Buc, E. Fox, and R. Garnett, eds.), Dec. 2019, pp. 10552-10563. Vancouver, Canada.


The Boolean Satisfiability (SAT) problem is the canonical NP-complete problem and is fundamental to computer science, with a wide array of applications in planning, verification, and theorem proving. Developing and evaluating practical SAT solvers relies on extensive empirical testing on a set of real-world benchmark formulas. However, the availability of such real-world SAT formulas is limited. While these benchmark formulas can be augmented with synthetically generated ones, existing approaches for doing so are heavily hand-crafted and fail to simultaneously capture a wide range of characteristics exhibited by real-world SAT instances. In this work, we present G2SAT, the first deep generative framework that learns to generate SAT formulas from a given set of input formulas. Our key insight is that SAT formulas can be transformed into latent bipartite graph representations which we model using a specialized deep generative neural network. We show that G2SAT can generate SAT formulas that closely resemble given real-world SAT instances, as measured by both graph metrics and SAT solver behavior. Further, we show that our synthetic SAT formulas could be used to improve SAT solver performance on real-world benchmarks, which opens up new opportunities for the continued development of SAT solvers and a deeper understanding of their performance.

BibTeX entry:

   author = {Jiaxuan You and Haoze Wu and Clark Barrett and Raghuram
	Ramanujan and Jure Leskovec},
   editor = {H. Wallach and H. Larochelle and A. Beygelzimer and F.
	d'Alch{\'e}-Buc and E. Fox and R. Garnett},
   title = {{G2SAT}: Learning to Generate {SAT} Formulas},
   booktitle = {Advances in Neural Information Processing Systems 32
	(NeurIPS '19)},
   pages = {10552--10563},
   publisher = {Curran Associates, Inc.},
   month = dec,
   year = {2019},
   note = {Vancouver, Canada},
   url = {http://theory.stanford.edu/~barrett/pubs/YWB+19.pdf}

(This webpage was created with bibtex2web.)