A decentralized federated learning method based on dynamic graphs and multi-criteria decision making for mobile networks

Document Type : Original Article

Authors

1 Phd candidate, Computer Engineering Department, Yazd University, Yazd, Iran

2 Associate Professor, Computer Engineering Department, Yazd University, Yazd, Iran

3 Assistant Professor, Computer Engineering Department, Yazd University, Yazd, Iran

Abstract

Federated learning, as a new approach in distributed machine learning, enables training of artificial intelligence models without the need to transfer raw data. However, the classical federated learning architecture depends on a central server, which is not available in many real-world applications such as mobile networks and the Internet of Things. On the other hand, data heterogeneity, node mobility and communication limitations create fundamental challenges in implementing such systems. In this study, a new method for decentralized federated learning is proposed. In this approach each node updates the model solely based on its local communications and without dependence on a central server. A dynamic graph based on node encounters is constructed to model the communication structure and effective nodes in the aggregation process are selected using the computation of the graph dominance set. To accurately weight the selected models, a multi-criteria decision-making method is used that considers criteria such as accuracy, data volume, and node score. In addition, the similarity between models is used the graph combination process. The proposed method is evaluated on three benchmark datasets MNIST, FASHION-MNIST, and CIFAR10 and is compared with two baseline methods, peer-to-peer FedAvg and WAFL. The accuracy results of the models in two environments with 50 and 100 nodes yielding values of 0.964, 0.952, 0.771, 0.682, 0.424 and 0.399 respectively, indicate that the proposed method achieves superior performance. These findings demonstrate the capability of the proposed approach for application in dynamic and mobile environments with a large number of clients and without centralized infrastructure, where it can overcome the scalability issues and client drift problem in decentralized federated learning.

Keywords

Main Subjects

References

[1]   B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Proc. Artif. Intell. Statist. (AISTATS), 2017, pp. 1273–1282.
[2]   P.Kairouz, H.B.McMahan et al., “Advances and Open Problems in federated Learning,” Found. Trends Mach. Learn.,vol.14, no.1, 2021.
[3]   J. Wang, Q. Liu, H. Liang, G. Joshi, and H. V. Poor, “Tackling the objective inconsistency problem in heterogeneous federated optimization,” Adv. Neural Inf. Process. Syst., vol. 33, pp. 7611–7623, 2020.
[4]   T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, and V. Smith, “Federated optimization in heterogeneous networks,” Proc. Mach. Learn. Syst., vol. 2, pp. 429–450, 2020.
[5]   J. Qi, Q. Zhou, L. Lei, and K. Zheng, “Federated reinforcement learning: Techniques, applications, and open challenges,” arXiv:2108.11887, 2021.
[6]   S. Kornblith, M. Norouzi, H. Lee, and G. Hinton, “Similarity of neural network representations revisited,” in Proc. Int. Conf. Mach. Learn. (ICML), 2019, pp. 3519–3529.
[7]   A. Karras et al., “Peer to peer federated learning: Towards decentralized machine learning on edge devices,” in Proc. SEEDA-CECNSM, 2022, pp. 1–9.
[8]   J. Qingzhe, “Decentralized P2P Federated Learning on Ad-hoc Like Networks with Non-IID Dataset,” M.S. thesis, Dept. Comput. Sci., Univ. Ottawa, Ottawa, ON, Canada, 2022.
[9]   Z. Qin et al., “BlockDFL: A blockchain-based fully decentralized peer-to-peer federated learning framework,” in Proc. ACM Web Conf. (WWW), 2024, pp. 2914–2925.
[10]     M. Mao et al., “SCAMA: A Smart-Contract-Driven Asynchronous Model Aggregation Framework for Decentralized Federated Learning,” in Proc. IEEE Int. Conf. High Perform. Smart Comput. (HPSC), 2025, pp. 102–107.
[11]     R. Chen et al., “FLock: Robust and Privacy-Preserving Federated Learning based on Practical Blockchain State Channels,” in Proc. ACM Web Conf. (WWW), 2025, pp. 884–895.
[12]     T. Shannon et al., “Exploring Heterogeneity in Federated Learning,” in Proc. Int. Conf. Wireless Artif. Intell. Comput. Syst. Appl., 2025, pp. 242–254.
[13]     S. Wang et al., “Towards Communication-Efficient Decentralized Federated Graph Learning over Non-IID Data,” arXiv:2509.08409, 2025.
[14]     H. Hegedűs, I. Hegedűs, A. Benczúr, and M. Jelasity, “Wireless Ad Hoc Federated Learning: A Fully Distributed Cooperative Machine Learning,” in Proc. IEEE International Conference on Distributed Computing Systems Workshops (ICDCSW), 2019, pp. 148–153.
[15]     V. Hegiste, T. Legler, and M. Ruskowski, “Towards robust federated image classification: An empirical study of weight selection strategies in manufacturing,” in Proc. Int. Conf. Federated Learn. Technol. Appl. (FLTA), 2024, pp. 55–62.
[16]     Y. Dutta, A. Sikder, and B. Palit, “Federated Learning for Deforestation Detection: A Distributed Approach with Satellite Imagery,” arXiv:2509.13631, 2025.
[17]     B. Büyüktaş, G. Sumbul, and B. Demir, “A multi-modal federated learning framework for remote sensing image classification,” arXiv:2503.10262, 2025.
[18]     O. Odeyomi et al., “Online Decentralized Federated Multi-task Learning With Trustworthiness in Cyber-Physical Systems,” arXiv:2509.00992, 2025.
[19]     M. Pavan and M. Pelillo, “Dominant sets and pairwise clustering,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 29, no. 1, pp. 167–172, Jan. 2007.
[20]     Y. Li et al., “ModelDiff: Testing Based DNN Similarity Comparison for Model Reuse Detection,” in Proc. ACM SIGSOFT Int. Symp. Softw. Test. Anal. (ISSTA), Jul. 2021, pp. 139–151.
[21]     M. Raghu et al., “SVCCA: Singular vector canonical correlation analysis for deep learning dynamics and interpretability,” Adv. Neural Inf. Process. Syst., vol. 30, 2017.
[22]     S. Arisdakessian, et al., "A Two-Level Dirichlet Framework for Heterogeneous Federated Network," in IEEE Transactions on Network Science and Engineering, vol. 13, pp.1599-1615, 2026.
[23] LeCun, Y., Cortes, C., & Burges, C. “The MNIST Database of Handwritten Digits”.1998.
[24] Xiao, H., Rasul, K., & Vollgraf, R.. "Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms". arXiv:1708.07747, 2017.
[25] Krizhevsky, A.“Learning Multiple Layers of Features from Tiny Images”. Technical Report, University of Toronto,CIFAR-10 and CIFAR-100 datasets, 2009.
Applied and basic Machine intelligence research
Volume 4, Issue 1
May 2026
Pages 183-197

Files

Share

How to cite

Statistics