Experience replay with cohesive-subgraph awareness for continual graph learning in IoT✩

Zhenzhen Xie; Qi Luo; Yan Huang; Yongqi Yin; Jiaqi Zhang; Junjie Pang

doi:10.1016/j.dcan.2025.08.010

›› 2026, Vol. 12 ›› Issue (3) :417 -440. DOI: 10.1016/j.dcan.2025.08.010

research-article

Experience replay with cohesive-subgraph awareness for continual graph learning in IoT^✩

Author information +

^a School of Computer Science and Technology, Shandong University , Shandong 266000, China

^b College of Computing and Software Engineering, Kennesaw State University , Atlanta, USA

^c Department of Computer Technology, Changchun University of Technology , Changchun 130012, China

^d College of Computer Science and Technology, Qingdao University , Shandong 266000, China

^* E-mail addresses: xiezz21@sdu.edu.cn (Z. Xie),

luoqi2018@mail.sdu.edu.cn (Q. Luo),

yan.huang@kennesaw.edu (Y. Huang),

yqy@mail.sdu.edu.cn (Y. Yin),

2202303117@stu.ccut.edu.cn (J. Zhang),

pangjj@qdu.edu.cn (J. Pang).

Zhenzhen Xie received her M.S. and Ph.D. degrees in Computer Science from Jilin University, China, in 2014 and 2021, respectively. She is currently an associate researcher in the School of Computer Science and Technology, Shandong University. Her research areas are edge computing, the Internet of Things, and federated learning.

Qi Luo received the B.S. degree in computer science from Northeastern University, Qinhuangdao, China, in 2015, and the M.E. and Ph.D. degrees from Shandong University, China, in 2018 and 2022, respectively. He is currently a professor with the School of Computer Science and Technology, Shandong University. His research interests include graph mining and analysis.

Yan Huang is currently an assistant professor in the Department of Software Engineering and Game Development at Kennesaw State University (KSU). Dr. Huang received his Ph.D. degree in the Department of Computing Science at Georgia State University. He is broadly interested in privacy and security, with particular emphasis on deep learning aided privacy protection solutions and cybersecurity challenges in the IoT environment.

Yongqi Yin received the B.Eng. degree in 2022 and the M.Eng. degree in 2025 from Shandong University, China. His research interests include graph computing and vector retrieval.

Jiaqi Zhang received the B.S. degree in 2023 in computer science and technology from the Department of Computer Science and Technology, Changchun University of Technology, Changchun, China, where she is currently working toward the M.S. degree in artificial intelligence. Her research interests include graph mining and analysis.

Junjie Pang received his M.S. and Ph.D. degrees in the Department of Computer Science at Jilin University in 2013 and 2017, respectively. He is currently an assistant professor at Qingdao University. Dr. Pang is the recipient of the Initiative Postdoctoral Program of Shandong Province. His research interests include federated learning, reinforcement learning, and next-generation networking.

Show less

History +

PDF (1419KB)

Abstract

In the Internet-of-Things (IoT) scenarios, Continual Graph Learning (CGL) has become a key technique for capturing the complex relational structures underlying diverse applications including sensor networks, road systems and biomedical networks. However, the structural changes in these evolving graphs introduce instability, making catastrophic forgetting a primary challenge for CGL. Experience replay is currently a promising method, as it strikes a balance between new and old knowledge. It also provides CGL models with a human-like memory capability. However, prior work rarely leverages the graph’s intrinsic properties to proactively identify the critical patterns hiding in the evolving graphs. To this end, we propose a unified framework that integrates cohesion-subgraph awareness into existing CGL mechanisms. We propose a novel cohesive structure-aware experience replay framework that leverages intrinsic graph properties, such as 𝑘-core and 𝑘-truss metrics, to guide the selection of representative historical nodes for replay. Unlike conventional replay strategies that rely on random sampling or task-driven node selection, our approach systematically identifies structurally significant nodes that encapsulate the evolving patterns of streaming graphs. By integrating these cohesive subgraph properties into the experience replay process, our method effectively preserves critical historical knowledge while adapting to new graph structures with low computational overhead. The experimental results demonstrate that our method consistently outperforms existing replay strategies in mitigating catastrophic forgetting and maintaining classification performance. On the PubMed dataset, our 𝑘-core-based replay strategy improves the 𝐹1 score by 3.7% compared to random sampling, while reducing training time by up to 85% compared to full retraining. Similarly, on the Cora dataset, our approach achieves a 98.3% 𝐹1 score, surpassing baseline methods by 4.5%.

Keywords

Graph neural networks / Continual learning / Experience replay / Cohesive subgraph

Cite this article

Download citation ▾

Zhenzhen Xie, Qi Luo, Yan Huang, Yongqi Yin, Jiaqi Zhang, Junjie Pang. Experience replay with cohesive-subgraph awareness for continual graph learning in IoT^✩. , 2026, 12 (3) : 417-440 DOI:10.1016/j.dcan.2025.08.010

登录浏览全文

4963

注册一个新账户忘记密码

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported in part by the Young Scientists Fund of the Natural Science Foundation of Shandong Province (ZR2022QF134).

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	G. Wu, H. Wang, Y. Liu, Y. He, Graphstgan: situation understanding network of slow—fast high maneuvering targets for maritime monitor services of iot data, Digit. Commun. Netw. 10 (3) (2024) 620-630.

[2]	Y. Li, S. Xie, Z. Wan, H. Lv, H. Song, Z. Lv, Graph—powered learning methods in the Internet of things: a survey, Mach. Learn. Appl. 11 (2023) 100441.

[3]	L.P. Rachakonda, M. Siddula, V. Sathya, A comprehensive study on iot privacy and security challenges with focus on spectrum sharing in next—generation networks (5g/6g/beyond), High—Confid. Comput. 4 (2) (2024) 100220.

[4]	F. Zhou, C. Cao, Overcoming catastrophic forgetting in graph neural networks with experience replay, in: Proceedings of the AAAI Conference on Artificial Intelligence, 2021, pp. 4714-4722.

[5]	X. Zhang, D. Song, D. Tao, Ricci curvature—based graph sparsification for continual graph representation learning, IEEE Trans. Neural Netw. Learn. Syst. 35 (12) (2024) 17398-17410.

[6]	L. Sang, H. Yuan, Y. Huang, Y. Zhang, Graph structure learning for robust recommendation, Tsinghua Sci. Technol. 30 (4) (2025) 1617-1635.

[7]	T.N. Kipf, M. Welling, Semi—supervised classification with graph convolutional networks, https://doi.org/10.48550/arXiv.1609.02907.

[8]	C. Zheng, B. Zong, W. Cheng, D. Song, J. Ni, W. Yu, H. Chen, W. Wang, Robust graph representation learning via neural sparsification, in: Proceedings of the 37th International Conference on Machine Learning, PMLR, 2020, pp. 11458-11468.

[9]	Y. Mo, H.T. Shen, X. Zhu, Unsupervised multi—view graph representation learning with dual weight—net, Inf. Fusion 114 (2025) 102669.

[10]	Z. Xing, Y. Ye, R. Song, Y. Teng, Z. Li, J. Liu, Sample feature enhancement model based on heterogeneous graph representation learning for few—shot relation classification, Inf. Sci. 690 (2025) 121583.

[11]	W.L. Hamilton, Z. Ying, J. Leskovec, Inductive representation learning on large graphs, in: Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, Long Beach, CA, USA, 2017.

[12]	A.A. Rusu, N.C. Rabinowitz, G. Desjardins, H. Soyer, J. Kirkpatrick, K. Kavukcuoglu, R. Pascanu, R. Hadsell, Progressive neural networks, https://doi.org/10.48550/arXiv.1606.04671.

[13]	L. Du, Y. Wang, G. Song, Z. Lu, J. Wang, Dynamic network embedding: an extended approach for skip—gram based network embedding, in: Proceedings of the Twenty—Seventh International Joint Conference on Artificial Intelligence, IJCAI 2018, July 13—19, 2018, Stockholm, Sweden, 2018.

[14]	G.H. Nguyen, J.B. Lee, R.A. Rossi, N.K. Ahmed, E. Koh, S. Kim, Continuous—time dynamic network embeddings, in: Companion of the Web Conference 2018 on the Web Conference 2018, WWW, 2018.

[15]	K. Zhang, G. Sang, J. Cheng, Z. Liu, Y. Zhang, Negative sampling strategy based on multi—hop neighbors for graph representation learning, Expert Syst. Appl. 263 (2025) 125688.

[16]	T. Bilot, N.E. Madhoun, K.A. Agha, A. Zouaoui, A survey on malware detection with graph representation learning, ACM Comput. Surv. 56 (11) (2024) 278.

[17]	L. Yang, C. Chatelain, S. Adam, Dynamic graph representation learning with neural networks: a survey, IEEE Access 12 (2024) 43460-43484.

[18]	D. Chen, S. Zheng, M. Xu, Z. Zhu, Y. Zhao, Signn: a spike—induced graph neural network for dynamic graph representation learning, Pattern Recognit. 158 (2025) 111026.

[19]	Y. Xu, Z. Peng, B. Shi, X. Hua, B. Dong, Learning dynamic graph representations through timespan view contrasts, Neural Netw. 176 (2024) 106384.

[20]	H. Yao, C. Zhang, Z. Yao, C.L.P. Chen, J. Hu, A recurrent graph neural network for inductive representation learning on dynamic graphs, Pattern Recognit. 154 (2024) 110577.

[21]	L. Wang, X. Zhang, H. Su, J. Zhu, A comprehensive survey of continual learning: theory, method and application, IEEE Trans. Pattern Anal. Mach. Intell. 46 (8) (2024) 5362-5383.

[22]	M. De Lange, R. Aljundi, M. Masana, S. Parisot, X. Jia, A. Leonardis, G. Slabaugh, T. Tuytelaars, A continual learning survey: defying forgetting in classification tasks, IEEE Trans. Pattern Anal. Mach. Intell. 44 (7) (2022) 3366-3385.

[23]	J. Kirkpatrick, R. Pascanu, N. Rabinowitz, J. Veness, G. Desjardins, A.A. Rusu, K. Milan, J. Quan, T. Ramalho, A. Grabska—Barwinska, D. Hassabis, C. Clopath, D. Kumaran, R. Hadsell, Overcoming catastrophic forgetting in neural networks, Proc. Natl. Acad. Sci. U.S.A. 114 (13) (2017) 3521-3526.

[24]	F. Zenke, B. Poole, S. Ganguli, Continual learning through synaptic intelligence, in: Proceedings of the 34th International Conference on Machine Learning, PMLR, 2017, pp. 3987-3995.

[25]	P. Li, L. Wang, W. Wu, F. Zhou, B. Wang, Q. Wu, Graph neural network—based scheduling for multi—uav—enabled communications in D2D networks, Digit. Commun. Netw. 10 (1) (2024) 45-52.

[26]	Z. He, Y. Li, D. Seo, Z. Cai, Fedcpd: addressing label distribution skew in federated learning with class proxy decoupling and proxy regularization, Inf. Fusion 110 (2024) 102481.

[27]	Z. Xiong, W. Li, Z. Cai, Federated generative model on multi—source heterogeneous data in IoT, in: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 37, 2023, pp. 10537-10545.

[28]	R. Aljundi, F. Babiloni, M. Elhoseiny, M. Rohrbach, T. Tuytelaars, Memory aware synapses: learning what (not) to forget, in: Computer Vision — ECCV 2018 — 15th European Conference, Munich, Germany, September 8—14, 2018, Proceedings, Part III, 2018.

[29]	A.S. Zacarias, L.A. Alexandre, Sena—cnn: overcoming catastrophic forgetting in convolutional neural networks by selective network augmentation, in: Artificial Neural Networks in Pattern Recognition — 8th IAPR TC3 Workshop, ANNPR, 2018.

[30]	T. Zhao, A. Hanjalic, M. Khosla, AGALE: a graph—aware continual learning evaluation framework, Trans. Mach. Learn. Res. (2024).

[31]	B. Tang, D.S. Matteson, Graph—based continual learning, in: 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3—7, 2021, 2021.

[32]	H. Bo, R. McConville, J. Hong, W. Liu, Ego—graph replay based continual learning for misinformation engagement prediction, in: International Joint Conference on Neural Networks, IJCNN, 2022.

[33]	C. Niu, G. Pang, L. Chen, Graph continual learning with debiased lossless memory replay, in: ECAI 2024 — 27th European Conference on Artificial Intelligence, Santiago de Compostela, Spain — Including 13th Conference on Prestigious Applications of Intelligent Systems PAIS, 2024.

[34]

Y. Fan, Y. Wang, P. Zhu, Q. Hu, Dynamic sub—graph distillation for robust semi—supervised continual learning, in: Thirty—Eighth AAAI Conference on Artificial Intelligence, AAAI 2024, Thirty—Sixth Conference on Innovative Applications of Artificial Intelligence, IAAI 2024, Fourteenth Symposium on Educational Advances in Artificial Intelligence, EAAI, 2024.