Contraction-Guided Adaptive Partitioning for Reachability Analysis of Neural Network Controlled Systems

Abstract

In this paper, we present a contraction-guided adaptive partitioning algorithm for improving interval-valued robust reachable set estimates in a nonlinear feedback loop with a neural network controller and disturbances. Based on an estimate of the contraction rate of over-approximated intervals, the algorithm chooses when and where to partition. Then, by leveraging a decoupling of the neural network verification step and reachability partitioning layers, the algorithm can provide accuracy improvements for little computational cost. This approach is applicable with any sufficiently accurate open-loop interval-valued reachability estimation technique and any method for bounding the input-output behavior of a neural network. Using contraction-based robustness analysis, we provide guarantees of the algorithm's performance with mixed monotone reachability. Finally, we demonstrate the algorithm's performance through several numerical simulations and compare it with existing methods in the literature. In particular, we report a sizable improvement in the accuracy of reachable set estimation in a fraction of the runtime as compared to state-of-the-art methods.

References in corpus (7)

• Efficient Neural Network Robustness Certification with General Activation Functions
• Automatic Perturbation Analysis for Scalable Certified Robustness and Beyond
• Non-Euclidean Contraction Theory for Robust Nonlinear Stability
• Open- and Closed-Loop Neural Network Verification using Polynomial Zonotopes
• Verification of Neural-Network Control Systems by Integrating Taylor Models and Zonotopes
• Fast Neural Network Verification via Shadow Prices
• Safety Verification of Neural Feedback Systems Based on Constrained Zonotopes

Cited by in corpus (2)

• Non-Euclidean Contraction Analysis of Continuous-Time Neural Networks
• Robust Model Predictive Control Exploiting Monotonicity Properties