Verisig 2.0: Verification of Neural Network Controllers Using Taylor Model Preconditioning [chapter]

Radoslav Ivanov, Taylor Carpenter, James Weimer, Rajeev Alur, George Pappas, Insup Lee
2021 Lecture Notes in Computer Science  
AbstractThis paper presents Verisig 2.0, a verification tool for closed-loop systems with neural network (NN) controllers. We focus on NNs with tanh/sigmoid activations and develop a Taylor-model-based reachability algorithm through Taylor model preconditioning and shrink wrapping. Furthermore, we provide a parallelized implementation that allows Verisig 2.0 to efficiently handle larger NNs than existing tools can. We provide an extensive evaluation over 10 benchmarks and compare Verisig 2.0
more » ... inst three state-of-the-art verification tools. We show that Verisig 2.0 is both more accurate and faster, achieving speed-ups of up to 21x and 268x against different tools, respectively.
doi:10.1007/978-3-030-81685-8_11 fatcat:ljm43ujxmzhd5dnxzavga7soqm