Representation Then Augmentation: Wide Graph Clustering Network With Multi-Order Filter Fusion and Double-Level Contrastive Learning

Youqing Wang; Tianxiang Zhao; Mingliang Cui; Junbin Gao; Li Liang; Jipeng Guo

doi:10.1109/JAS.2025.125564

IEEE/CAA Journal of Automatica Sinica

JCR Impact Factor: 19.2, Top 1 (SCI Q1)

CiteScore: 28.2, Top 1% (Q1)
Google Scholar h5-index: 95， TOP 5

Turn off MathJax

Article Contents

Article Navigation > IEEE/CAA Journal of Automatica Sinica > 2026 > In Press, Accepted Manuscript

Y. Wang, T. Zhao, M. Cui, J. Gao, L. Liang, and J. Guo, “Representation then augmentation: Wide graph clustering network with multi-order filter fusion and double-level contrastive learning,” IEEE/CAA J. Autom. Sinica, vol. 13, no. 2, pp. 1–15, Feb. 2026. doi: 10.1109/JAS.2025.125564

Citation:

Y. Wang, T. Zhao, M. Cui, J. Gao, L. Liang, and J. Guo, “Representation then augmentation: Wide graph clustering network with multi-order filter fusion and double-level contrastive learning,” IEEE/CAA J. Autom. Sinica, vol. 13, no. 2, pp. 1–15, Feb. 2026. doi: 10.1109/JAS.2025.125564

Citation:

Y. Wang, T. Zhao, M. Cui, J. Gao, L. Liang, and J. Guo, “Representation then augmentation: Wide graph clustering network with multi-order filter fusion and double-level contrastive learning,” IEEE/CAA J. Autom. Sinica, vol. 13, no. 2, pp. 1–15, Feb. 2026. doi: 10.1109/JAS.2025.125564

PDF( 7726 KB)

Representation Then Augmentation: Wide Graph Clustering Network With Multi-Order Filter Fusion and Double-Level Contrastive Learning

doi: 10.1109/JAS.2025.125564

Funds: This research was supported by the National Natural Science Foundation of China (62225303, 62403043, 62433004), the Beijing Natural Science Foundation (4244085), the Postdoctoral Fellowship Program of CPSF (GZC20230203) and the China Postdoctoral Science Foundation (2023M740201)

More Information

Author Bio:
Youqing Wang (Senior Member, IEEE) received the B.S. degree in mathematics from Shandong University in 2003, and the Ph.D. degree in control science and engineering from Tsinghua University in 2008. He worked chronologically at The Hong Kong University of Science and Technology, Hong Kong, China; the University of California at Santa Barbara USA, the University of Alberta, Canada; the Shandong University of Science and Technology, and the City University of Hong Kong, China. He is currently a Professor with the Beijing University of Chemical Technology. His research interests include industrial data analysis, machine learning, fault diagnosis, fault-tolerant control, and iterative learning control for chemical and biomedical processes. Dr. Wang was a recipient of several honors and awards, including the IET Fellow, the NSFC Distinguished Young Scientists Fund, the Journal of Process Control Survey Paper Prize, and the ADCHEM2015 Young Author Prize

Tianxiang Zhao received the B.S. degree in automation from Beijing University of Chemical Technology, in 2024. Currently, he is a master student degree in control science and engineering at the College of Information Science and Technology, Beijing University of Chemical Technology. His research interests include machine learning, graph clustering and large language models

Mingliang Cui received the B.S. degree in automation and the M.S. degree in control engineering from Shandong University of Science and Technology, in 2018 and 2021, respectively. He is currently a Ph.D. degree candidate in control science and engineering with the College of Information Science and Technology, Beijing University of Chemical Technology. His research interests include machine learning, industrial data mining, data-driven fault diagnosis, and process monitoring

Junbin Gao received the B.Sc degree in computational mathematics from Huazhong University of Science and Technology (HUST), China in 1982 and the Ph.D degree from Dalian University of Technology, in 1991. He is a Professor of big data analytics with the University of Sydney Business School, the University of Sydney Australia and was a Professor in Computer Science with the School of Computing and Mathematics, Charles Sturt University, Australia. He was a Senior lecturer, a Lecturer in Computer Science from 2001 to 2005 at the University of New England, Australia. From 1982 to 2001 he was an Associate Lecturer, Lecturer, Associate Professor, and Professor with Department of Mathematics, HUST. His main research interests include machine learning, data analytics, Bayesian learning and inference, and image analysis

Li Liang received the Ph.D degree in control science and engineering from the Beijing Institute of Technology, in 2020. She is currently an Associate Professor with the College of Information Science and Technology, Beijing University of Chemical Technology. Her research interests include differential games, multi-agent systems, multi-objective optimization and decision

Jipeng Guo (Member, IEEE) received the Ph.D. degree in control science and engineering from the Beijing University of Technology, in 2023. He is currently a Lecturer with the College of Information Science and Technology, Beijing University of Chemical Technology. He has published several papers in top journals and conferences, such as the IEEE TPAMI/TIP/TNNLS/TBD/TCST/TIM, the ACM TKDD, the Neural Networks, Information Fusion and the AAAI/ICASSP. His current research interests include pattern recognition, machine learning, and clustering method
Corresponding author: Jipeng Guo, e-mail: guojipeng@buct.edu.cn
Received Date: 2025-03-19
Accepted Date: 2025-05-28

Available Online: 2026-01-31

Abstract

Abstract

Deep graph contrastive clustering has attracted widespread attentions due to its self-supervised representation learning paradigm and superior clustering performance. Although, two challenges emerge and result in high computational costs. Most existing contrastive methods adopt the data augmentation and then representation learning strategy, where representation learning with trainable graph convolution is coupled with complex and fixed data augmentation, inevitably limiting the efficiency and flexibility. The similarity metric between positive-negative sample pairs is complex and contrastive objective is partial, limiting the discriminability of representation learning. To solve these challenges, a novel wide graph clustering network (WGCN) adhering to representation and then augmentation framework is proposed, which mainly consists of multi-order filter fusion (MFF) and double-level contrastive learning (DCL) modules. Specifically, the MFF module integrates multi-order low-pass filters to extract smooth and multi-scale topological features, utilizing self-attention fusion to reduce redundancy and obtain comprehensive embedding representation. Further, the DCL module constructs two augmented views by the parallel parameter-unshared Siamese encoders rather than complex augmentations on graph. To achieve simple yet effective self-supervised learning, representation self-supervision and structural consistency oriented double-level contrastive loss is designed, where representation self-supervision maximizes the agreement between pairwise augmented embedding representations and structural consistency promotes the mutual information correlation between appending neighborhoods with similar semantics. Extensive experiments on six benchmark datasets demonstrate the superiority of the proposed WGCN, especially highlighting its time-saving characteristic. The code could be available in the https://github.com/TianxiangZhao0474/WGCN.git.

FullText(HTML)

References(47)

References

[1]	G. Fortino, F. Messina, D. Rosaci, and G. M. Sarnè, “ResIoT: An IoT social framework resilient to malicious activities,” IEEE/CAA J. Autom. Sinica, vol. 7, no. 5, pp. 1263–1278, 2020. doi: 10.1109/JAS.2020.1003330
[2]	Y. Tian, Y. Feng, X. Zhang, and C. Sun, “A fast clustering based evolutionary algorithm for super-large-scale sparse multi-objective optimization,” IEEE/CAA J. Autom. Sinica, vol. 10, no. 4, pp. 1048–1063, 2022.
[3]	Y. Ren, J. Pu, Z. Yang, J. Xu, G. Li, X. Pu, S. Y. Philip, and L. He, “Deep clustering: A comprehensive survey,” IEEE Trans. Neural Networks and Learning Systems, 2024, doi: 10.1109/TNNLS.2024.3403155.
[4]	Y. Qin, H. Wu, J. Zhao, and G. Feng, “Enforced block diagonal subspace clustering with closed form solution,” Pattern Recognition, vol. 130, p. 108791, 2022. doi: 10.1016/j.patcog.2022.108791
[5]	Y. Yang, Y. Sun, S. Wang, J. Guo, J. Gao, F. Ju, and B. Yin, “Graph neural networks with soft association between topology and attribute,” in Proc. AAAI, 2024, pp. 9260–9268.
[6]	Z. Chen, Z. Wu, S. Wang, and W. Guo, “Dual low-rank graph autoencoder for semantic and topological networks,” in Proc. AAAI, 2023, pp. 4191–4198.
[7]	Z. Chen, Z. Wu, Z. Lin, S. Wang, C. Plant, and W. Guo, “AGNN: Alternating graph-regularized neural networks to alleviate over-smoothing,” IEEE Trans. Neural Networks and Learning Systems, vol. 35, no. 10, pp. 13764–13776, 2024. doi: 10.1109/TNNLS.2023.3271623
[8]	G. Wan, Y. Tian, W. Huang, N. V. Chawla, and M. Ye, “S3GCL: Spectral, swift, spatial graph contrastive learning,” in Proc. ICML, 2024.
[9]	G. Wan, W. Huang, and M. Ye, “Federated graph learning under domain shift with generalizable prototypes,” in Proc. AAAI, 2024, pp. 15429–15437.
[10]	J. Zhao, J. Guo, Y. Sun, J. Gao, S. Wang, and B. Yin, “Adaptive graph convolutional clustering network with optimal probabilistic graph,” Neural Networks, vol. 156, pp. 271–284, 2022. doi: 10.1016/j.neunet.2022.09.017
[11]	X. Wu, Z. Ren, and F. R. Yu, “Parameter-free shifted laplacian reconstruction for multiple kernel clustering,” IEEE/CAA J. Autom. Sinica, vol. 11, no. 4, pp. 1072–1074, 2023.
[12]	S. Wang, X. Lin, Z. Fang, S. Du, and G. Xiao, “Contrastive consensus graph learning for multi-view clustering,” IEEE/CAA J. Autom. Sinica, vol. 9, no. 11, pp. 2027–2030, 2022. doi: 10.1109/JAS.2022.105959
[13]	K. Liang, L. Meng, M. Liu, Y. Liu, W. Tu, S. Wang, S. Zhou, X. Liu, F. Sun, and K. He, “A survey of knowledge graph reasoning on graph types: Static, dynamic, and multi-modal,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 46, no. 12, pp. 9456–9478, 2024. doi: 10.1109/TPAMI.2024.3417451
[14]	Z. Wei, H. Zhao, Z. Li, X. Bu, Y. Chen, X. Zhang, Y. Lv, and F.-Y. Wang, “STGSA: A novel spatial-temporal graph synchronous aggregation model for traffic prediction,” IEEE/CAA J. Autom. Sinica, vol. 10, no. 1, pp. 226–238, 2023. doi: 10.1109/JAS.2023.123033
[15]	T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” in Proc. ICLR, 2017.
[16]	Z. Wu, S. Pan, F. Chen, G. Long, C. Zhang, and S. Y. Philip, “A comprehensive survey on graph neural networks,” IEEE Trans. Neural Networks and Learning Systems, vol. 32, no. 1, pp. 4–24, 2021. doi: 10.1109/TNNLS.2020.2978386
[17]	X. Hong, T. Zhang, Z. Cui, and J. Yang, “Variational gridded graph convolution network for node classification,” IEEE/CAA J. Autom. Sinica, vol. 8, no. 10, pp. 1697–1708, 2021. doi: 10.1109/JAS.2021.1004201
[18]	S. Wang, J. Yang, J. Yao, Y. Bai, and W. Zhu, “An overview of advanced deep graph node clustering,” IEEE Trans. Comput. Social Systems, vol. 11, no. 1, pp. 1302–1314, 2024. doi: 10.1109/TCSS.2023.3242145
[19]	Y. Liu, J. Xia, S. Zhou, X. Yang, K. Liang, C. Fan, Y. Zhuang, S. Z. Li, X. Liu, and K. He, “A survey of deep graph clustering: Taxonomy, challenge, application, and open resource,” arXiv preprint arXiv: 2211.12875, 2022.
[20]	G. Du, L. Zhou, Z. Li, L. Wang, and K. Lü, “Neighbor-aware deep multi-view clustering via graph convolutional network,” Information Fusion, vol. 93, pp. 330–343, 2023. doi: 10.1016/j.inffus.2023.01.001
[21]	P. Veličković, G. Cucurull, A. Casanova, A. Romero, P. Lio, and Y. Bengio, “Graph attention networks,” in Proc. ICLR, 2018.
[22]	C. Wang, S. Pan, R. Hu, G. Long, J. Jiang, and C. Zhang, “Attributed graph clustering: A deep attentional embedding approach,” in Proc. IJCAI, 2019, pp. 3670–3676.
[23]	D. Bo, X. Wang, C. Shi, M. Zhu, E. Lu, and P. Cui, “Structural deep clustering network,” in Proc. WWW, 2020, pp. 1400–1410.
[24]	W. Tu, S. Zhou, X. Liu, X. Guo, Z. Cai, E. Zhu, and J. Cheng, “Deep fusion clustering network,” in Proc. AAAI, 2021, pp. 9978–9987.
[25]	J. Cheng, Q. Wang, Z. Tao, D. Xie, and Q. Gao, “Multi-view attribute graph convolution networks for clustering,” in Proc. IJCAI, 2021, pp. 2973–2979.
[26]	Z. Chen, L. Fu, J. Yao, W. Guo, C. Plant, and S. Wang, “Learnable graph convolutional network and feature fusion for multi-view learning,” Information Fusion, vol. 95, pp. 109–119, 2023. doi: 10.1016/j.inffus.2023.02.013
[27]	S. Pan, R. Hu, G. Long, J. Jiang, L. Yao, and C. Zhang, “Adversarially regularized graph autoencoder for graph embedding,” in Proc. IJCAI, 2018, pp. 2609–2615.
[28]	Z. Tao, H. Liu, J. Li, Z. Wang, and Y. Fu, “Adversarial graph embedding for ensemble clustering,” in Proc. IJCAI, 2019, pp. 3562–3568.
[29]	Y. Yang, F. Ju, Y. Sun, J. Gao, and B. Yin, “Adversarially regularized joint structured clustering network,” Information Sciences, vol. 615, pp. 136–151, 2022. doi: 10.1016/j.ins.2022.09.066
[30]	H. Zhao, X. Yang, Z. Wang, E. Yang, and C. Deng, “Graph debiased contrastive learning with joint representation clustering.” in Proc. IJCAI, 2021, pp. 3434–3440.
[31]	K. Hassani and A. H. Khasahmadi, “Contrastive multi-view representation learning on graphs,” in Proc. ICML, 2020, pp. 4116–4126.
[32]	W. Xia, Q. Wang, Q. Gao, M. Yang, and X. Gao, “Self-consistent contrastive attributed graph clustering with pseudo-label prompt,” IEEE Trans. Multimedia, vol. 25, pp. 6665–6677, 2022.
[33]	Y. Liu, W. Tu, S. Zhou, X. Liu, L. Song, X. Yang, and E. Zhu, “Deep graph clustering via dual correlation reduction,” in Proc. AAAI, 2022, pp. 7603–7611.
[34]	X. Yang, C. Tan, Y. Liu, K. Liang, S. Wang, S. Zhou, J. Xia, S. Z. Li, X. Liu, and E. Zhu, “Convert: Contrastive graph clustering with reliable augmentation,” in Proc. ACM MM, 2023, pp. 319–327.
[35]	X. Shen, D. Sun, S. Pan, X. Zhou, and L. T. Yang, “Neighbor contrastive learning on learnable graph augmentation,” in Proc. AAAI, 2023, pp. 9782–9791.
[36]	X. Yan, K. Deng, Q. Zou, Z. Tian, and H. Yu, “Self-cumulative contrastive graph clustering,” IEEE/CAA J. Autom. Sinica, 2025, doi: 10.1109/JAS.2024.125025.
[37]	T. N. Kipf and M. Welling, “Variational graph auto-encoders,” in NIPS, 2016.
[38]	G. Huo, Y. Zhang, J. Gao, B. Wang, Y. Hu, and B. Yin, “CaEGCN: Cross-attention fusion based enhanced graph convolutional network for clustering,” IEEE Trans. Knowledge and Data Engineering, vol. 35, no. 4, pp. 3471–3483, 2023. doi: 10.1109/TKDE.2021.3125020
[39]	W. Xia, Q. Gao, M. Yang, and X. Gao, “Self-supervised contrastive attributed graph clustering,” arXiv preprint arXiv: 2110.08264, 2021.
[40]	X. Yang, E. Min, K. LIANG, Y. Liu, S. Wang, H. Wu, X. Liu, E. Zhu et al., “GraphLearner: Graph node clustering with fully learnable augmentation,” in Proc. ACM MM, 2024, pp. 5517–5526.
[41]	X. Yang, Y. Liu, S. Zhou, S. Wang, W. Tu, Q. Zheng, X. Liu, L. Fang, and E. Zhu, “Cluster-guided contrastive graph clustering network,” in Proc. AAAI, 2023, pp. 10834–10842.
[42]	Y. Liu, X. Yang, S. Zhou, X. Liu, S. Wang, K. Liang, W. Tu, and L. Li, “Simple contrastive graph clustering,” IEEE Trans. Neural Networks and Learning Systems, vol. 35, no. 10, pp. 13 789–13 800, 2024. doi: 10.1109/TNNLS.2023.3271871
[43]	W. Xia, S. Wang, M. Yang, Q. Gao, J. Han, and X. Gao, “Multi-view graph embedding clustering network: Joint self-supervision and block diagonal representation,” Neural Networks, vol. 145, pp. 1–9, 2022. doi: 10.1016/j.neunet.2021.10.006
[44]	G. Cui, J. Zhou, C. Yang, and Z. Liu, “Adaptive graph encoder for attributed graph embedding,” in Proc. SIGKDD, 2020, pp. 976–985.
[45]	W. Jin, X. Liu, X. Zhao, Y. Ma, N. Shah, and J. Tang, “Automated self-supervised learning for graphs,” arXiv preprint arXiv: 2106.05470, 2021.
[46]	N. Lee, J. Lee, and C. Park, “Augmentation-free self-supervised learning on graphs,” in Proc. AAAI, 2022, pp. 7372–7380.
[47]	L. van der Maaten and G. Hinton, “Visualizing data using t-SNE,” J. Machine Learning Research, vol. 9, no. 86, pp. 2579–2605, 2008.

Supplements(0)

Cited By

Proportional views

Proportional views

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

Figures(7) / Tables(9)

Get Citation

PDF

XML

Article Metrics

Article views (41) PDF downloads(1)

Representation Then Augmentation: Wide Graph Clustering Network With Multi-Order Filter Fusion and Double-Level Contrastive Learning

doi: 10.1109/JAS.2025.125564

Abstract

References

Proportional views

Catalog

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Export File

Citation

Format

Content