2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2017)
Honolulu, Hawaii, USA
July 21, 2017 to July 26, 2017
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CVPR.2017.17
Given a state-of-the-art deep neural network classifier, we show the existence of a universal (image-agnostic) and very small perturbation vector that causes natural images to be misclassified with high probability. We propose a systematic algorithm for computing universal perturbations, and show that state-of-the-art deep neural networks are highly vulnerable to such perturbations, albeit being quasi-imperceptible to the human eye. We further empirically analyze these universal perturbations and show, in particular, that they generalize very well across neural networks. The surprising existence of universal perturbations reveals important geometric correlations among the high-dimensional decision boundary of classifiers. It further outlines potential security breaches with the existence of single directions in the input space that adversaries can possibly exploit to break a classifier on most natural images.
image classification, neural nets, probability
S. Moosavi-Dezfooli, A. Fawzi, O. Fawzi and P. Frossard, "Universal Adversarial Perturbations," 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, USA, 2017, pp. 86-94.