Foveation-based Mechanisms Alleviate Adversarial Examples

TitleFoveation-based Mechanisms Alleviate Adversarial Examples
Publication TypeCBMM Memos
Year of Publication2016
AuthorsLuo, Y, Boix, X, Roig, G, Poggio, TA, Zhao, Q
Number044
Date Published01/2016
Publication LanguageEnglish
Abstract

We show that adversarial examples, i.e., the visually imperceptible perturbations that result in Convolutional Neural Networks (CNNs) fail, can be alleviated with a mechanism based on foveations---applying the CNN in different image regions. To see this, first, we report results in ImageNet that lead to a revision of the hypothesis that adversarial perturbations are a consequence of CNNs acting as a linear classifier: CNNs act locally linearly to changes in the image regions with objects recognized by the CNN, and in other regions the CNN may act non-linearly. Then, we corroborate that when the neural responses are linear, applying the foveation mechanism to the adversarial example tends to significantly reduce the effect of the perturbation. This is because, hypothetically, the CNNs for ImageNet are robust to changes of scale and translation of the object produced by the foveation, but this property does not generalize to transformations of the perturbation. As a result, the accuracy after a foveation is almost the same as the accuracy of the CNN without the adversarial perturbation, even if the adversarial perturbation is calculated taking into account a foveation.

arXiv

arXiv:1511.06292

DSpace@MIT

http://hdl.handle.net/1721.1/100981

Download: 

CBMM Memo No: 

044

Research Area: 

CBMM Relationship: 

  • CBMM Funded