
The runway object classifier under study is considerably more vulnerable to noise perturbations than to brightness or contrast perturbations, indicating the need for further robustness improvements.


coremsg

assessing-the-robustness-of-a-runway-object-classifier-for-safe-aircraft-taxiing


Assessing the Robustness of a Runway Object Classifier for Safe Aircraft Taxiing


title_rewrite


This paper proposes a unified framework for both qualitative and quantitative safety verification of DNN-controlled systems by leveraging neural barrier certificates.


unifying-qualitative-and-quantitative-safety-verification-of-deep-neural-network-controlled-systems


Unifying Qualitative and Quantitative Safety Verification of Deep Neural Network-Controlled Systems
