Explaining Away Attacks Against Neural Networks
Machine Learning
2020-03-13 v1 Cryptography and Security
Computer Vision and Pattern Recognition
Machine Learning
Abstract
We investigate the problem of identifying adversarial attacks on image-based neural networks. We present intriguing experimental results showing significant discrepancies between the explanations generated for the predictions of a model on clean and adversarial data. Utilizing this intuition, we propose a framework which can identify whether a given input is adversarial based on the explanations given by the model. Code for our experiments can be found here: https://github.com/seansaito/Explaining-Away-Attacks-Against-Neural-Networks.
Cite
@article{arxiv.2003.05748,
title = {Explaining Away Attacks Against Neural Networks},
author = {Sean Saito and Jin Wang},
journal= {arXiv preprint arXiv:2003.05748},
year = {2020}
}
Comments
2 pages, 2 figures; Accepted at MLSys 2020 First Workshop on Secure and Resilient Autonomy