V-PSC: A perturbation-based causative attack against DL classifiers' supply chain in VANET

Yi Zeng; Meikang Qiu; Jingqi Niu; Yanxin Long; Jian Xiong; Meiqin Liu

doi:10.1109/CSE/EUC.2019.00026

V-PSC: A perturbation-based causative attack against DL classifiers' supply chain in VANET

Yi Zeng
, Meikang Qiu
, Jingqi Niu
, Yanxin Long
, Jian Xiong
, Meiqin Liu

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

DL Based classifiers can attain a higher accuracy with less storage requirement, which suits perfectly with the VANET. However, it has been proved that DL models suffer from crafted perturbation data, a small amount of such can misguide the classifier, thus backdoors can be created for malicious reasons. This paper studies such a causative attack in the VANET. We present a perturbation-based causative attack which targets at the supply chain of DL classifiers in the VANET. We first train a classifier using VANET simulated data which meets the standard accuracy for identifying malicious traffic in the VANET. Then, we elaborate on the effectiveness of our presented attack scheme on this pre-trained classifier. We also explore some feasible approaches to ease the outcome brought by our attack. Experimental results show that the scheme can cause the target DL model a 10.52% drop in accuracy.

Original language	English
Title of host publication	Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019
Editors	Meikang Qiu
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	86-91
Number of pages	6
ISBN (Electronic)	9781728116631
DOIs	https://doi.org/10.1109/CSE/EUC.2019.00026
State	Published - Aug 2019
Externally published	Yes
Event	22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019 - New York, United States Duration: 1 Aug 2019 → 3 Aug 2019

Publication series

Name	Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019

Conference

Conference	22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019
Country/Territory	United States
City	New York
Period	1/08/19 → 3/08/19

Keywords

Causative Attack
Deep Learning
Perturbation
VANET

Access to Document

10.1109/CSE/EUC.2019.00026

Cite this

Zeng, Y., Qiu, M., Niu, J., Long, Y., Xiong, J., & Liu, M. (2019). V-PSC: A perturbation-based causative attack against DL classifiers' supply chain in VANET. In M. Qiu (Ed.), Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019 (pp. 86-91). Article 8919585 (Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CSE/EUC.2019.00026

Zeng, Yi ; Qiu, Meikang ; Niu, Jingqi et al. / V-PSC : A perturbation-based causative attack against DL classifiers' supply chain in VANET. Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019. editor / Meikang Qiu. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 86-91 (Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019).

@inproceedings{baefe2a4605e4aa6a38d36aecdad34e5,

title = "V-PSC: A perturbation-based causative attack against DL classifiers' supply chain in VANET",

abstract = "DL Based classifiers can attain a higher accuracy with less storage requirement, which suits perfectly with the VANET. However, it has been proved that DL models suffer from crafted perturbation data, a small amount of such can misguide the classifier, thus backdoors can be created for malicious reasons. This paper studies such a causative attack in the VANET. We present a perturbation-based causative attack which targets at the supply chain of DL classifiers in the VANET. We first train a classifier using VANET simulated data which meets the standard accuracy for identifying malicious traffic in the VANET. Then, we elaborate on the effectiveness of our presented attack scheme on this pre-trained classifier. We also explore some feasible approaches to ease the outcome brought by our attack. Experimental results show that the scheme can cause the target DL model a 10.52\% drop in accuracy.",

keywords = "Causative Attack, Deep Learning, Perturbation, VANET",

author = "Yi Zeng and Meikang Qiu and Jingqi Niu and Yanxin Long and Jian Xiong and Meiqin Liu",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019 ; Conference date: 01-08-2019 Through 03-08-2019",

year = "2019",

month = aug,

doi = "10.1109/CSE/EUC.2019.00026",

language = "英语",

series = "Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "86--91",

editor = "Meikang Qiu",

booktitle = "Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019",

}

Zeng, Y, Qiu, M, Niu, J, Long, Y, Xiong, J & Liu, M 2019, V-PSC: A perturbation-based causative attack against DL classifiers' supply chain in VANET. in M Qiu (ed.), Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019., 8919585, Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019, Institute of Electrical and Electronics Engineers Inc., pp. 86-91, 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019, New York, United States, 1/08/19. https://doi.org/10.1109/CSE/EUC.2019.00026

V-PSC: A perturbation-based causative attack against DL classifiers' supply chain in VANET. / Zeng, Yi; Qiu, Meikang; Niu, Jingqi et al.
Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019. ed. / Meikang Qiu. Institute of Electrical and Electronics Engineers Inc., 2019. p. 86-91 8919585 (Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - V-PSC

T2 - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019

AU - Zeng, Yi

AU - Qiu, Meikang

AU - Niu, Jingqi

AU - Long, Yanxin

AU - Xiong, Jian

AU - Liu, Meiqin

PY - 2019/8

Y1 - 2019/8

N2 - DL Based classifiers can attain a higher accuracy with less storage requirement, which suits perfectly with the VANET. However, it has been proved that DL models suffer from crafted perturbation data, a small amount of such can misguide the classifier, thus backdoors can be created for malicious reasons. This paper studies such a causative attack in the VANET. We present a perturbation-based causative attack which targets at the supply chain of DL classifiers in the VANET. We first train a classifier using VANET simulated data which meets the standard accuracy for identifying malicious traffic in the VANET. Then, we elaborate on the effectiveness of our presented attack scheme on this pre-trained classifier. We also explore some feasible approaches to ease the outcome brought by our attack. Experimental results show that the scheme can cause the target DL model a 10.52% drop in accuracy.

AB - DL Based classifiers can attain a higher accuracy with less storage requirement, which suits perfectly with the VANET. However, it has been proved that DL models suffer from crafted perturbation data, a small amount of such can misguide the classifier, thus backdoors can be created for malicious reasons. This paper studies such a causative attack in the VANET. We present a perturbation-based causative attack which targets at the supply chain of DL classifiers in the VANET. We first train a classifier using VANET simulated data which meets the standard accuracy for identifying malicious traffic in the VANET. Then, we elaborate on the effectiveness of our presented attack scheme on this pre-trained classifier. We also explore some feasible approaches to ease the outcome brought by our attack. Experimental results show that the scheme can cause the target DL model a 10.52% drop in accuracy.

KW - Causative Attack

KW - Deep Learning

KW - Perturbation

KW - VANET

UR - https://www.scopus.com/pages/publications/85077059747

U2 - 10.1109/CSE/EUC.2019.00026

DO - 10.1109/CSE/EUC.2019.00026

M3 - 会议稿件

AN - SCOPUS:85077059747

T3 - Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019

SP - 86

EP - 91

BT - Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019

A2 - Qiu, Meikang

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 1 August 2019 through 3 August 2019

ER -

Zeng Y, Qiu M, Niu J, Long Y, Xiong J, Liu M. V-PSC: A perturbation-based causative attack against DL classifiers' supply chain in VANET. In Qiu M, editor, Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 86-91. 8919585. (Proceedings - 22nd IEEE International Conference on Computational Science and Engineering and 17th IEEE International Conference on Embedded and Ubiquitous Computing, CSE/EUC 2019). doi: 10.1109/CSE/EUC.2019.00026

V-PSC: A perturbation-based causative attack against DL classifiers' supply chain in VANET

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this