Input Validation for Neural Networks via Runtime Local Robustness Verification

Abstract

Local robustness verification can verify that a neural network is robust wrt. any perturbation to a specific input within a certain distance. We call this distance Robustness Radius. We observe that the robustness radii of correctly classified inputs are much larger than that of misclassified inputs which include adversarial examples, especially those from strong adversarial attacks. Another observation is that the robustness radii of correctly classified inputs often follow a normal distribution. Based on these two observations, we propose to validate inputs for neural networks via runtime local robustness verification. Experiments show that our approach can protect neural networks from adversarial examples and improve their accuracies.

References in corpus (5)

• ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Models
• Evaluating the Robustness of Neural Networks: An Extreme Value Theory Approach
• Finding Anomalous Periodic Time Series: An Application to Catalogs of Periodic Variable Stars
• Adversarial Examples, Uncertainty, and Transfer Testing Robustness in Gaussian Process Hybrid Deep Networks
• Outside the Box: Abstraction-Based Monitoring of Neural Networks