Fault Sneaking Attack: a Stealthy Framework for Misleading Deep Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhao, Pu, Wang, Siyue, Gongye, Cheng, Wang, Yanzhi, Fei, Yunsi, Lin, Xue
Format: Preprint
Published: 2019
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908432335896576
author Zhao, Pu
Wang, Siyue
Gongye, Cheng
Wang, Yanzhi
Fei, Yunsi
Lin, Xue
author_facet Zhao, Pu
Wang, Siyue
Gongye, Cheng
Wang, Yanzhi
Fei, Yunsi
Lin, Xue
contents Despite the great achievements of deep neural networks (DNNs), the vulnerability of state-of-the-art DNNs raises security concerns of DNNs in many application domains requiring high reliability.We propose the fault sneaking attack on DNNs, where the adversary aims to misclassify certain input images into any target labels by modifying the DNN parameters. We apply ADMM (alternating direction method of multipliers) for solving the optimization problem of the fault sneaking attack with two constraints: 1) the classification of the other images should be unchanged and 2) the parameter modifications should be minimized. Specifically, the first constraint requires us not only to inject designated faults (misclassifications), but also to hide the faults for stealthy or sneaking considerations by maintaining model accuracy. The second constraint requires us to minimize the parameter modifications (using L0 norm to measure the number of modifications and L2 norm to measure the magnitude of modifications). Comprehensive experimental evaluation demonstrates that the proposed framework can inject multiple sneaking faults without losing the overall test accuracy performance.
format Preprint
id arxiv_https___arxiv_org_abs_1905_12032
institution arXiv
publishDate 2019
record_format arxiv
spellingShingle Fault Sneaking Attack: a Stealthy Framework for Misleading Deep Neural Networks
Zhao, Pu
Wang, Siyue
Gongye, Cheng
Wang, Yanzhi
Fei, Yunsi
Lin, Xue
Machine Learning
Cryptography and Security
Computer Vision and Pattern Recognition
Despite the great achievements of deep neural networks (DNNs), the vulnerability of state-of-the-art DNNs raises security concerns of DNNs in many application domains requiring high reliability.We propose the fault sneaking attack on DNNs, where the adversary aims to misclassify certain input images into any target labels by modifying the DNN parameters. We apply ADMM (alternating direction method of multipliers) for solving the optimization problem of the fault sneaking attack with two constraints: 1) the classification of the other images should be unchanged and 2) the parameter modifications should be minimized. Specifically, the first constraint requires us not only to inject designated faults (misclassifications), but also to hide the faults for stealthy or sneaking considerations by maintaining model accuracy. The second constraint requires us to minimize the parameter modifications (using L0 norm to measure the number of modifications and L2 norm to measure the magnitude of modifications). Comprehensive experimental evaluation demonstrates that the proposed framework can inject multiple sneaking faults without losing the overall test accuracy performance.
title Fault Sneaking Attack: a Stealthy Framework for Misleading Deep Neural Networks
topic Machine Learning
Cryptography and Security
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/1905.12032