Proof-Driven Clause Learning in Neural Network Verification
Isac, Omri, Refaeli, Idan, Wu, Haoze, Barrett, Clark, Katz, Guy
–arXiv.org Artificial Intelligence
The widespread adoption of deep neural networks (DNNs) requires efficient techniques for safety verification. Existing methods struggle to scale to real-world DNNs, and tremendous efforts are being put into improving their scalability. In this work, we propose an approach for improving the scalability of DNN verifiers using Conflict-Driven Clause Learning (CDCL) -- an approach that has proven highly successful in SAT and SMT solving. We present a novel algorithm for deriving conflict clauses using UNSAT proofs, and propose several optimizations for expediting it. Our approach allows a modular integration of SAT solvers and DNN verifiers, and we implement it on top of an interface designed for this purpose. The evaluation of our implementation over several benchmarks suggests a 2X--3X improvement over a similar approach, with specific cases outperforming the state of the art.
arXiv.org Artificial Intelligence
Mar-15-2025
- Country:
- North America > United States (0.46)
- Genre:
- Research Report (1.00)
- Industry:
- Government > Regional Government (0.46)
- Information Technology
- Robotics & Automation (0.46)
- Security & Privacy (0.46)
- Transportation (0.68)
- Technology: