Backdoor Attacks on Graph Neural Networks Trained with Data Augmentation

被引：0

作者：

Yashiki, Shingo ^{[1
]}

Takahashi, Chako ^{[2
]}

Suzuki, Koutarou ^{[1
]}

机构：

[1] Toyohashi Univ Technol, Dept Comp Sci & Engn, Toyohashi 4418580, Japan

[2] Yamagata Univ, Grad Sch Sci & Engn, Yonezawa 9928510, Japan

来源：

IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES | 2024年 / E107A卷 / 03期

关键词：

data augmentation; backdoor attacks; graph neural networks; graph classification;

D O I：

10.1587/transfun.2023CIL0007

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper investigates the effects of backdoor attacks on graph neural networks (GNNs) trained through simple data augmentation by modifying the edges of the graph in graph classification. The numerical results show that GNNs trained with data augmentation remain vulnerable to backdoor attacks and may even be more vulnerable to such attacks than GNNs without data augmentation.

引用

页码：355 / 358

页数：4

共 50 条

[1] Backdoor Attacks to Graph Neural Networks
Zhang, Zaixi
Jia, Jinyuan
Wang, Binghui
Gong, Neil Zhenqiang
[J]. PROCEEDINGS OF THE 26TH ACM SYMPOSIUM ON ACCESS CONTROL MODELS AND TECHNOLOGIES, SACMAT 2021, 2021, : 15 - 26
[2] Watermarking Graph Neural Networks based on Backdoor Attacks
Xu, Jing
Koffas, Stefanos
Ersoy, Oguzhan
Picek, Stjepan
[J]. 2023 IEEE 8TH EUROPEAN SYMPOSIUM ON SECURITY AND PRIVACY, EUROS&P, 2023, : 1179 - 1197
[3] More is Better (Mostly): On the Backdoor Attacks in Federated Graph Neural Networks
Xu, Jing
Wang, Rui
Koffas, Stefanos
Liang, Kaitai
Picek, Stjepan
[J]. PROCEEDINGS OF THE 38TH ANNUAL COMPUTER SECURITY APPLICATIONS CONFERENCE, ACSAC 2022, 2022, : 684 - 698
[4] Multi-target label backdoor attacks on graph neural networks
Wang, Kaiyang
Deng, Huaxin
Xu, Yijia
Liu, Zhonglin
Fang, Yong
[J]. PATTERN RECOGNITION, 2024, 152
[5] Data Augmentation for Graph Neural Networks
Zhao, Tong
Liu, Yozen
Neves, Leonardo
Woodford, Oliver
Jiang, Meng
Shah, Neil
[J]. THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2021, 35 : 11015 - 11023
[6] Backdoor smoothing: Demystifying backdoor attacks on deep neural networks
Grosse, Kathrin
Lee, Taesung
Biggio, Battista
Park, Youngja
Backes, Michael
Molloy, Ian
[J]. Computers and Security, 2022, 120
[7] Backdoor smoothing: Demystifying backdoor attacks on deep neural networks
Grosse, Kathrin
Lee, Taesung
Biggio, Battista
Park, Youngja
Backes, Michael
Molloy, Ian
[J]. COMPUTERS & SECURITY, 2022, 120
[8] Inconspicuous Data Augmentation Based Backdoor Attack on Deep Neural Networks
Xu, Chaohui
Liu, Wenye
Zheng, Yue
Wang, Si
Chang, Chip-Hong
[J]. 2022 IEEE 35TH INTERNATIONAL SYSTEM-ON-CHIP CONFERENCE (IEEE SOCC 2022), 2022, : 237 - 242
[9] Rationalizing Graph Neural Networks with Data Augmentation
Liu, Gang
Inae, Eric
Luo, Tengfei
Jiang, Meng
[J]. ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA, 2024, 18 (04)
[10] Adversarial Attacks on Neural Networks for Graph Data
Zuegner, Daniel
Akbarnejad, Amir
Guennemann, Stephan
[J]. PROCEEDINGS OF THE TWENTY-EIGHTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2019, : 6246 - 6250

← 1 2 3 4 5 →