Motif-Backdoor: Rethinking the Backdoor Attack on Graph Neural Networks via Motifs
Open accessAbstract
Graph neural network (GNN) with a powerful representation capability has been widely applied to various areas. Recent works have exposed that GNN is vulnerable to the backdoor attack, i.e., models trained with maliciously crafted training samples are easily fooled by patched samples. Most of the proposed studies launch the backdoor attack using a trigger that is either the randomly generated subgraph [e.g., erdős-rényi backdoor (ER-B)] for less computational burden or the gradient-based generative subgraph [e.g., graph trojaning attack (GTA)] to enable a more effective attack. However, the interpretation of how is the trigger structure and the effect of the backdoor attack related has been overlooked in the current literature. Motifs, recurrent and statistically significant subgraphs in graphs, contain rich structure information. In this article, we are rethinking the trigger from the perspective of motifs and propose a motif-based backdoor attack, denoted as Motif-Backdoor. It contributes from three aspects. First, Interpretation, it provides an in-depth explanation for backdoor effectiveness by the validity of the trigger structure from motifs, leading to some novel insights, e.g., using subgraphs that appear less frequently in the graph as the trigger can achieve better attack performance. Second, Effectiveness, Motif-Backdoor reaches the state-of-the-art (SOTA) attack performance in both black-box and defensive scenarios. Third, Efficiency, based on the graph motif distribution, Motif-Backdoor can quickly obtain an effective trigger structure without target model feedback or subgraph model generation. Extensive experimental results show that Motif-Backdoor realizes the SOTA performance on three popular models and four public datasets compared with five baselines, e.g., Motif-Backdoor improves the attack success rate (ASR) by 14.73% compared with baselines on average. In addition, under a possible defense, Motif-Backdoor still implements satisfying performance, highlighting the requirement of defenses against backdoor attacks on GNNs. The datasets and code are available athttps://github.com/Seaocn/Motif-Backdoor.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Existing works show reasonable defense performance on label-flipping backdoor attack and trigger-based backdoor attack.
“The existing works show reasonable defense performance on label-flipping backdoor attack [186], [187] and trigger-based backdoor attack [188], [189].”
From Data and model poisoning backdoor attacks on wireless federated learning, and the defense mechanisms: A comprehensive survey (2024) · cites this paper as [188]
Cites
Centrality in social networks conceptual clarification
Freeman, L. C. (1978) · Social Networks · doi:10.1016/0378-8733(78)90021-7 · Cited by 17,062 · see citing works
node2vec: Scalable Feature Learning for Networks
Grover, A., & Leskovec, J. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939754 · Cited by 10,926 · see citing works
DeepWalk: online learning of social representations
Perozzi, B., Al-Rfou, R., & Skiena, S. (2014) · Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/2623330.2623732 · Cited by 10,123 · see citing works
Network motifs: simple building blocks of complex networks.
Milo, R., Shen-Orr, S., Itzkovitz, S., Kashtan, N., Chklovskii, D., & Alon, U. (2002) · Science · doi:10.1126/science.298.5594.824 · Cited by 7,368 · see citing works
Network motifs: theory and experimental approaches
Alon, U. (2007) · Nature Reviews Genetics · doi:10.1038/nrg2102 · Cited by 3,279 · see citing works
The Network Data Repository with Interactive Graph Analytics and Visualization
Rossi, R. A., & Ahmed, N. K. (2015) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v29i1.9277 · Cited by 2,619 · see citing works
Weisfeiler-Lehman Graph Kernels
Shervashidze, N., Schweitzer, P., van Leeuwen, E. J., Mehlhorn, K., & Borgwardt, K. M. (2010) · Max Planck Digital Library · Cited by 1,608 · see citing works
Protein function prediction via graph kernels
Borgwardt, K. M., Ong, C. S., Schönauer, S., Vishwanathan, S. V. N., Smola, A. J., & Kriegel, H.-P. (2005) · Bioinformatics · doi:10.1093/bioinformatics/bti1007 · Cited by 1,103 · see citing works
Adversarial Attacks on Neural Networks for Graph Data
Zügner, D., Akbarnejad, A., & Günnemann, S. (2018) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3219819.3220078 · Cited by 997 · see citing works
Local Higher-Order Graph Clustering
Yin, H., Benson, A. R., Leskovec, J., & Gleich, D. F. (2017) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3097983.3098069 · Cited by 620 · see citing works
Adversarial Examples for Graph Data: Deep Insights into Attack and Defense
Wu, H., Wang, C., Tyshetskiy, Y., Docherty, A., Lü, K., & Zhu, L. (2019) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2019/669 · Cited by 467 · see citing works
Smart Contract Vulnerability Detection using Graph Neural Network
Zhuang, Y., Liu, Z., Qian, P., Liu, Q., Wang, X., & He, Q. (2020) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2020/454 · Cited by 429 · see citing works
Backdoor Attacks to Graph Neural Networks
Zhang, Z., Jia, J., Wang, B., & Gong, N. Z. (2021) · Proceedings of the 26th ACM Symposium on Access Control Models and Technologies · doi:10.1145/3450569.3463560 · Cited by 220 · see citing works
A combinatorial approach to graphlet counting
Hočevar, T., & Demšar, J. (2013) · Bioinformatics · doi:10.1093/bioinformatics/btt717 · Cited by 217 · see citing works
Graph convolutional network soft sensor for process quality prediction
Jia, M., Xu, D., Yang, T., Liu, Y., & Yao, Y. (2023) · Journal of Process Control · doi:10.1016/j.jprocont.2023.01.010 · Cited by 157 · see citing works
Deep Autoencoder Thermography for Defect Detection of Carbon Fiber Composites
Liu, K., Zheng, M., Liu, Y., et al. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3172902 · Cited by 120 · see citing works
Graph Convolutional Networks with Motif-based Attention
Lee, J. B., Rossi, R. A., Kong, X.-N., Kim, S., Koh, E., & Rao, A. B. (2019) · International Conference on Information and Knowledge Management · doi:10.1145/3357384.3357880 · Cited by 110 · see citing works
Adversarial Attack on Large Scale Graph
Li, J., Xie, T., Chen, L., Xie, F., He, X., & Zheng, Z. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3078755 · Cited by 81 · see citing works
Explainability-based Backdoor Attacks Against Graph Neural Networks
Xu, J., Xue, M., & Picek, S. (2021) · ACM Workshop on Wireless Security and Machine Learning (WiseML) · doi:10.1145/3468218.3469046 · Cited by 79 · see citing works
Zola, F., Segurola-Gil, L., Bruse, J. L., Galar, M., & Orduna-Urrutia, R. (2022) · Computers & Security · doi:10.1016/j.cose.2022.102632 · Cited by 70 · see citing works
Graph Adversarial Attack via Rewiring
Ma, Y., Wang, S., Derr, T., Wu, L., & Tang, J. (2021) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3447548.3467416 · Cited by 69 · see citing works
MODEL: Motif-Based Deep Feature Learning for Link Prediction
Wang, L., Ren, J. F., Xu, B., Li, J., Luo, W., & Xia, F. (2020) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2019.2962819 · Cited by 66 · see citing works
Topology-Guided Graph Learning for Process Fault Diagnosis
Jia, M., Hu, J., Liu, Y., et al. (2023) · Industrial & Engineering Chemistry Research · doi:10.1021/acs.iecr.2c03628 · Cited by 64 · see citing works
Zheng, H., Chen, Z., Du, T., Zhang, X., Cheng, Y., Ji, S., Wang, J., Yu, Y., & Chen, J. (2022) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1145/3510003.3510123 · Cited by 63 · see citing works
Self-citationNode, Motif and Subgraph: Leveraging Network Functional Blocks Through Structural Convolution
Yang, C., Liu, M., Zheng, V. W., et al. (2018) · IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) · doi:10.1109/asonam.2018.8508729 · Cited by 58 · see citing works
Graph stream classification using labeled and unlabeled graphs
Pan, S., Xingquan Zhu, A., Zhang, C., & Yu, P. S. (2013) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde.2013.6544842 · Cited by 50 · see citing works
A Holistic Feature Selection Method for Enhanced Short-term Load Forecasting of Power System
Jiang, B., Liu, Y., Geng, H., Wang, Y., Zeng, H., & Ding, J. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3219499 · Cited by 46 · see citing works
Time-aware Gradient Attack on Dynamic Network Link Prediction
Chen, J., Zhang, J., Chen, Z., et al. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3110580 · Cited by 40 · see citing works
Self-citationPredicting Customer Value with Social Relationships via Motif-based Graph Attention Networks
Piao, J., Zhang, G., Xu, F., Chen, Z., & Li, Y. (2021) · International World Wide Web Conference (WWW) · doi:10.1145/3442381.3449849 · Cited by 36 · see citing works
Motif Enhanced Recommendation over Heterogeneous Information Network
Zhao, H., Zhou, Y., Song, Y., et al. (2019) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3357384.3358134 · Cited by 31 · see citing works
Motif-based Convolutional Neural Network on GraphsPreprint
Sankar, A., Zhang, X., & Chang, K. C. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.05697 · Cited by 29 · see citing works
Hu, W. S., Xia, X., Ding, X. H., Zhang, X., Zhong, K., & Zhang, H. (2022) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2022.3220260 · Cited by 29 · see citing works
GRIP-GAN: An Attack-Free Defense Through General Robust Inverse Perturbation
Zheng, H., Chen, J., Du, H., Zhu, W., Ji, S., & Zhang, X. (2021) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2021.3124337 · Cited by 28 · see citing works
Self-citationmotif2vec: Motif Aware Node Representation Learning for Heterogeneous Networks
Dareddy, M. R., Das, M., & Yang, H. (2019) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata47090.2019.9005670 · Cited by 25 · see citing works
GraphAttacker: A General Multi-Task Graph Attack Framework
Chen, J., Zhang, D., Ming, Z., Huang, K., Jiang, W., & Cui, C. (2021) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2021.3127557 · Cited by 23 · see citing works
Self-citationShao, P. H., Yang, Y., Xu, S., et al. (2021) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3473911 · Cited by 21 · see citing works
EGC2: Enhanced graph classification with easy graph compression
Chen, J., Xiong, H., Zheng, H., et al. (2023) · Information Sciences · doi:10.1016/j.ins.2023.01.114 · Cited by 19 · see citing works
Self-citationMTGK: Multi-source cross-network node classification via transferable graph knowledge
Yang, H., He, H. X., Zhang, W., et al. (2022) · Information Sciences · doi:10.1016/j.ins.2022.01.007 · Cited by 17 · see citing works
Backdoor Attack of Graph Neural Networks Based on Subgraph Trigger
Sheng, Y., Chen, R., Cai, G., & Kuang, L. (2021) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-030-92638-0_17 · Cited by 12 · see citing works
Influence maximization in social networks with privacy protection
Zhang, X.-J., Wang, J., Ma, X., Ma, C., Kan, J.-Q., & Zhang, H. (2022) · Physica A Statistical Mechanics and its Applications · doi:10.1016/j.physa.2022.128179 · Cited by 11 · see citing works
MotifExplainer: a Motif-based Graph Neural Network ExplainerPreprint
Yu, Z., & Gao, H. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2202.00519 · Cited by 7 · see citing works
Hierarchical Structure-Feature Aware Graph Neural Network for Node Classification
Yao, W., Guo, K., Hou, Y., & Li, X. (2022) · IEEE Access · doi:10.1109/access.2022.3164691 · Cited by 6 · see citing works
Machine learning prediction of network dynamics with privacy protection
Xia, X., Su, Y., Lü, L., Zhang, X., Lai, Y., & Zhang, H. (2022) · Physical Review Research · doi:10.1103/physrevresearch.4.043076 · Cited by 5 · see citing works
Motif-Aware Adversarial Graph Representation Learning
Zhao, M. J., Zhang, Y., Xia, X., et al. (2022) · IEEE Access · doi:10.1109/access.2022.3144233 · Cited by 3 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Wan, Y., Qu, Y., Ni, W., Xiang, Y., Gao, L., & Hossain, E. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3361451 · Cited by 152 · see citing works
“The existing works show reasonable defense performance on label-flipping backdoor attack [186], [187] and trigger-based backdoor attack [188], [189].”
Background
- Authors
- Zheng, H., Xiong, H., Chen, J., Ma, H., & Huang, G.
- Affiliations
- Zhejiang University of Technology (CN)
- Venue
- IEEE Transactions on Computational Social Systems
- Published
- 2024-04
- Volume
- 11
- Issue
- 2
- Pages
- 2479-2493
- ISSN
- 2329-924X (ISSN-L)
- 2373-7476
- OpenAlex
- W4367146727
- Cited by
- 41
- References
- 44
- Citations by year
- 2026: 3 · 2025: 23 · 2024: 13 · 2023: 2
- FWCI
- 3.63
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Computational Social Systems (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- NSFC (62072406)
- Zhejiang Provincial Natural Science Foundation (LDQ23F020001)
- Chinese National Key Laboratory of Science and Technology on Information System Security (61421110502)
- National Key Research and Development Projects of China (2018AAA0100801)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Artificial Intelligence › Advanced Graph Neural Networks
- Keywords
- backdoor attacks, graph neural networks, graph motifs, trigger design, adversarial training, black-box attacks, backdoor defense
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
