2 papers
cs.CR2026
Differential Zonotopes for Verifying Global Robustness of DNNs
Anagha Athavale, Samuel Teuber, Matteo Maffei +3
The robustness of deep neural networks (DNNs) is critical in security-sensitive applications, where small input perturbations should not alter model predictions. This property is c…
cs.LO2024
Verifying Global Two-Safety Properties in Neural Networks with Confidence
Anagha Athavale, Ezio Bartocci, Maria Christakis +3
We present the first automated verification technique for confidence-based 2-safety properties, such as global robustness and global fairness, in deep neural networks (DNNs). Our a…