Tag
The paper proposes an attack-agnostic robustness metric based on the spectral norm of the Fisher Information Matrix, providing theoretical bounds and scalable evaluation methods for deep neural networks.
This paper demonstrates that deep neural networks are catastrophically vulnerable to minimal sign-bit flips in parameters, introducing DNL and 1P-DNL methods to identify critical vulnerable parameters without data or optimization. The vulnerability spans multiple domains including image classification, object detection, instance segmentation, and language models, with practical implications for model security.