Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
126 tokens/sec
GPT-4o
47 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

G4SATBench: Benchmarking and Advancing SAT Solving with Graph Neural Networks (2309.16941v2)

Published 29 Sep 2023 in cs.LG

Abstract: Graph neural networks (GNNs) have recently emerged as a promising approach for solving the Boolean Satisfiability Problem (SAT), offering potential alternatives to traditional backtracking or local search SAT solvers. However, despite the growing volume of literature in this field, there remains a notable absence of a unified dataset and a fair benchmark to evaluate and compare existing approaches. To address this crucial gap, we present G4SATBench, the first benchmark study that establishes a comprehensive evaluation framework for GNN-based SAT solvers. In G4SATBench, we meticulously curate a large and diverse set of SAT datasets comprising 7 problems with 3 difficulty levels and benchmark a broad range of GNN models across various prediction tasks, training objectives, and inference algorithms. To explore the learning abilities and comprehend the strengths and limitations of GNN-based SAT solvers, we also compare their solving processes with the heuristics in search-based SAT solvers. Our empirical results provide valuable insights into the performance of GNN-based SAT solvers and further suggest that existing GNN models can effectively learn a solving strategy akin to greedy local search but struggle to learn backtracking search in the latent space. Our codebase is available at https://github.com/zhaoyu-li/G4SATBench.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (44)
  1. Learning to solve Circuit-SAT: An unsupervised differentiable approach. In International Conference on Learning Representations (ICLR), 2019a.
  2. PDP: A general neural framework for learning constraint satisfaction solvers. arXiv preprint arXiv:1903.01969, 2019b.
  3. Layer normalization. arXiv preprint arXiv:1607.06450, 2016.
  4. Improving stochastic local search for sat with a new probability distribution. In Theory and Applications of Satisfiability Testing–SAT 2010: 13th International Conference, SAT 2010, Edinburgh, UK, July 11-14, 2010. Proceedings 13, pp.  10–15. Springer, 2010.
  5. Handbook of Satisfiability, volume 185. IOS press, 2009.
  6. Cliques in random graphs. In Mathematical Proceedings of the Cambridge Philosophical Society, 1976.
  7. Graph neural networks and boolean satisfiability. arXiv preprint arXiv:1702.03592, 2017.
  8. Predicting propositional satisfiability via end-to-end learning. In AAAI Conference on Artificial Intelligence (AAAI), 2020.
  9. A simple framework for contrastive learning of visual representations. In International Conference on Machine Learning (ICML), 2020.
  10. Graph neural reasoning may fail in certifying boolean unsatisfiability. arXiv preprint arXiv:1909.11588, 2019.
  11. Experimental results on the crossover point in random 3-SAT. Artificial Intelligence, 1996.
  12. Augment with care: Contrastive learning for combinatorial problems. In International Conference on Machine Learning (ICML), 2022.
  13. Fast graph representation learning with pytorch geometric. arXiv preprint arXiv:1903.02428, 2019.
  14. Cadical, kissat, paracooba, plingeling and treengeling entering the sat competition 2020. SAT COMPETITION, 2020.
  15. A modularity-based random SAT instances generator. In International Joint Conference on Artificial Intelligence (IJCAI), 2015.
  16. Locality in random SAT instances. In International Joint Conference on Artificial Intelligence (IJCAI), 2017.
  17. Machine learning methods in solving the boolean satisfiability problem. arXiv preprint arXiv:2203.04755, 2022.
  18. Jesse Michael Han. Enhancing SAT solvers with glue variable predictions. arXiv preprint arXiv:2007.02559, 2020.
  19. Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In IEEE International Conference on Computer Vision (ICCV), 2015.
  20. Sean B Holden et al. Machine learning for automated theorem proving: Learning to solve SAT and QSAT. Foundations and Trends® in Machine Learning, 14(6):807–989, 2021.
  21. SATLIB: An online resource for research on SAT. Workshop on Satisfiability (SAT), 2000.
  22. Neural heuristics for SAT solving. arXiv preprint arXiv:2005.13406, 2020.
  23. Adam: A method for stochastic optimization. In International Conference on Learning Representations (ICLR), 2015.
  24. Semi-supervised classification with graph convolutional networks. In International Conference on Learning Representations (ICLR), 2017.
  25. Can q-learning with graph networks learn a generalizable branching heuristic for a SAT solver? In Advances in Neural Information Processing Systems (NeurIPS), 2020.
  26. Cnfgen: A generator of crafted benchmarks. In Theory and Applications of Satisfiability Testing (SAT), 2017.
  27. On eda-driven learning for sat solving. In Design Automation Conference (DAC), 2023.
  28. Gated graph sequence neural networks. In International Conference on Learning Representations (ICLR), 2016.
  29. NSNet: A general neural probabilistic framework for satisfiability problems. In Advances in Neural Information Processing Systems (NeurIPS), 2022.
  30. Rectified linear units improve restricted boltzmann machines. In International Conference on Machine Learning (ICML), 2010.
  31. Goal-aware neural SAT solver. In International Joint Conference on Neural Networks (IJCNN), 2022.
  32. Pytorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems (NeurIPS), 2019.
  33. A new method for solving hard satisfiability problems. In National Conference on Artificial Intelligence (AAAI), 1992.
  34. Guiding high-performance SAT solvers with unsat-core predictions. In Theory and Applications of Satisfiability Testing (SAT), 2019.
  35. Learning a SAT solver from single-bit supervision. In International Conference on Learning Representations (ICLR), 2019.
  36. Satformer: Transformer-based unsat core learning. In International Conference on Computer Aided Design (ICCAD), 2023.
  37. João P. Marques Silva and Karem A. Sakallah. GRASP: A search algorithm for propositional satisfiability. IEEE Transactions on Computers, 1999.
  38. Neurocomb: Improving SAT solving with graph neural networks. arXiv preprint arXiv:2110.14053, 2021.
  39. Drat-trim: Efficient checking and trimming using expressive clausal proofs. In Theory and Applications of Satisfiability Testing (SAT), 2014.
  40. On the domination number of a random graph. The Electronic Journal of Combinatorics, 2001.
  41. How powerful are graph neural networks? In International Conference on Learning Representations (ICLR), 2019.
  42. Addressing variable dependency in gnn-based SAT solving. arXiv preprint arXiv:2304.08738, 2023.
  43. Learning local search heuristics for boolean satisfiability. In Advances in Neural Information Processing Systems (NeurIPS), 2019.
  44. Nlocalsat: Boosting local search with solution prediction. In International Joint Conference on Artificial Intelligence (IJCAI), 2020.
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now