Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Dec 4;14(1):30222.
doi: 10.1038/s41598-024-79824-y.

DGHSA: derivative graph-based hypergraph structure attack

Yang Chen  1   2   3 , Zhonglin Ye  4   5 , Zhaoyang Wang  4   5 , Jingjing Lin  4   5 , Haixing Zhao  6   7
Affiliations

DGHSA: derivative graph-based hypergraph structure attack

Yang Chen et al. Sci Rep. .

Abstract

Hypergraph Neural Networks (HGNNs) have been significantly successful in higher-order tasks. However, recent study have shown that they are also vulnerable to adversarial attacks like Graph Neural Networks. Attackers fool HGNNs by modifying node links in hypergraphs. Existing adversarial attacks on HGNNs only consider feasibility in the targeted attack, and there is no discussion on the untargeted attack with higher practicality. To close this gap, we propose a derivative graph-based hypergraph attack, namely DGHSA, which focuses on reducing the global performance of HGNNs. Specifically, DGHSA consists of two models: candidate set generation and evaluation. The gradients of the incidence matrix are obtained by training HGNNs, and then the candidate set is obtained by modifying the hypergraph structure with the gradient rules. In the candidate set evaluation module, DGHSA uses the derivative graph metric to assess the impact of attacks on the similarity of candidate hypergraphs, and finally selects the candidate hypergraph with the worst node similarity as the optimal perturbation hypergraph. We have conducted extensive experiments on four commonly used datasets, and the results show that DGHSA can significantly degrade the performance of HGNNs on node classification tasks.

PubMed Disclaimer

Conflict of interest statement

Competing interests: The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
DGHSA overall framework. DGHAS consists of two modules: (a)–(c) are expressed as candidate set generation module, (d)–(e) are expressed as evaluation candidate set module.
Algorithm 1
Algorithm 1
Derivative graph-based hypergraph atructure attack (DGHSA).
Fig. 2
Fig. 2
The impact of budget on CSR. The first and second rows show the results for HGNN-KNN and HGNN-ε under the four datasets, respectively. Attack performance is positively related to attack budget.
Fig. 3
Fig. 3
CSR of DGHSA with different parameters K. The parameter K does not affect the performance of DGHSA.
Fig. 4
Fig. 4
CSR of DGHSA with different parameters ε. The parameter ε does not affect the performance of DGHSA.
Fig. 5
Fig. 5
CSR in different parameters n. The performance of DGHSA becomes better with the increase of n.
Fig. 6
Fig. 6
Original hypergraph (a) and modified hypergraph (b)–(e).
See this image and copyright information in PMC

References

    1. Li, R. et al. Graph signal processing, graph neural network and graph learning on biological data: A systematic review. IEEE Rev. Biomed. Eng.16, 109–135. 10.1109/RBME.2021.3122522 (2023). - PubMed
    1. Zhang, R., Zhang, Y., Lu, C. & Li, X. Unsupervised graph embedding via adaptive graph learning. IEEE Trans. Pattern Anal. Mach. Intell.45, 5329–5336. 10.1109/TPAMI.2022.3202158 (2023). - PubMed
    1. Ji, J., Jia, H., Ren, Y. & Lei, M. Supervised contrastive learning with structure inference for graph classification. IEEE Trans. Netw. Sci. Eng.10, 1684–1695. 10.1109/TNSE.2022.3233479 (2023).
    1. Wang, K. et al. Minority-weighted graph neural network for imbalanced node classification in social networks of internet of people. IEEE Internet Things J.10, 330–340. 10.1109/JIOT.2022.3200964 (2023).
    1. Lyu, Z. et al. Knowledge enhanced graph neural networks for explainable recommendation. IEEE Trans. Knowl. Data Eng.35, 4954–4968 (2023).

LinkOut - more resources