Federated Machine Learning Algorithm Based on Swarm Intelligence with Local Differential Privacy

• Júlia Almeida Luna UFMG
• Cauã Ferreira Sathler Siman UFMG
• Layane Garcia Pereira UFMG
• Thiago Lucas de Oliveira UFMG
• Eliezer Timoteo da Silva Sanhá UFMG
• Frederico Gadelha Guimarães UFMG

DOI: https://doi.org/10.5753/sbseg.2025.9805

Abstract

In federated learning, malicious agents may exploit different types of cyberattacks to manipulate the results of predictive models or to infer information about the training data. To mitigate such risks, this paper presents Fed-DPP-SO, a federated machine learning method that combines Swarm Intelligence and Local Differential Privacy. The approach aims to protect distributed data and hinder the extraction of sensitive information. The experiments conducted indicate that Fed-DP-PSO is a promising solution for training models in federated contexts with differential privacy, as its performance proved superior compared to the FedAvg method with DP-SGD.

References

Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., and Zhang, L. (2016). Deep learning with differential privacy. In Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security, CCS’16. ACM.

Freitas, D., Lopes, L. G., and Morgado-Dias, F. (2020). Particle swarm optimisation: A historical review up to the current developments. Entropy, 22(3).

Jagielski, M., Oprea, A., Biggio, B., Liu, C., Nita-Rotaru, C., and Li, B. (2018). Manipulating machine learning: Poisoning attacks and countermeasures for regression learning. In 2018 IEEE Symposium on Security and Privacy (SP), pages 19–35.

Kennedy, J. and Eberhart, R. (1995). Particle swarm optimization. In Proceedings of ICNN’95 - International Conference on Neural Networks, volume 4, pages 1942–1948 vol.4.

Leite, L., Santo, Y., Dalmazo, B., and Riker, A. (2024). Federated learning under attack: Improving gradient inversion for batch of images. In Anais do XXIV Simpósio Brasileiro de Segurança da Informação e de Sistemas Computacionais, pages 794–800, Porto Alegre, RS, Brasil. SBC.

Luzón, M. V., Rodríguez-Barroso, N., Argente-Garrido, A., Jiménez-López, D., Moyano, J. M., Del Ser, J., Ding, W., and Herrera, F. (2024). A tutorial on federated learning from theory to practice: Foundations, software frameworks, exemplary use cases, and selected trends. IEEE/CAA Journal of Automatica Sinica, 11(4):824–850.

Majeed, A. and Lee, S. (2021). Anonymization techniques for privacy preserving data publishing: A comprehensive survey. IEEE Access, 9:8512–8545.

McMahan, B., Moore, E., Ramage, D., Hampson, S., and y Arcas, B. A. (2017). Communication-efficient learning of deep networks from decentralized data. In Artificial intelligence and statistics, pages 1273–1282. PMLR.

Narayanan, A. and Shmatikov, V. (2008). Robust de-anonymization of large sparse datasets. In 2008 IEEE Symposium on Security and Privacy (sp 2008), pages 111–125. IEEE.

Nguyen, L. M., Hoang, T. N., and Chen, P.-Y. (2024). Federated Learning: Theory and Practice. Academic Press.

Pan, K., Ong, Y.-S., Gong, M., Li, H., Qin, A., and Gao, Y. (2024). Differential privacy in deep learning: A literature survey. Neurocomputing, 589:127663.

Park, S., Suh, Y., and Lee, J. (2021). Fedpso: Federated learning using particle swarm optimization to reduce communication costs. Sensors, 21(2).

Sanhá, E. T., Erazo-Costa, F. J., and Guimaraes, F. G. (2024). Algoritmo hıbrido de otimizaçao por enxame de partıculas para o aprendizado federado de redes neurais artificiais. In Anais do Congresso Brasileiro de Automática (CBA), pages 1–6.

Sanhá, E. (2024). Algoritmo híbrido de otimização por enxame de partículas para o aprendizado federado de redes neurais artificiais. Master’s thesis, Escola de Engenharia.

Silveira, M., Portela, A., Souza, M., Silva, D., Mesquita, M., Silva, D., Menezes, R., and Gomes, R. (2023). Aplicação de técnicas de encriptação e anonimização em nuvem para proteção de dados. In Anais do XXIII Simpósio Brasileiro de Segurança da Informação e de Sistemas Computacionais, pages 111–124, Porto Alegre, RS, Brasil. SBC.

Tatarakis, N. (2019). Differentially private federated learning. Master’s thesis, KTH, School of Electrical Engineering and Computer Science (EECS).

Yu, J., Moon, H., Chua, B.-L., and and, H. H. (2022). Hotel data privacy: strategies to reduce customers’ emotional violations, privacy concerns, and switching intention. Journal of Travel & Tourism Marketing, 39(2):215–227.

Zhang, C., Xie, Y., Bai, H., Yu, B., Li, W., and Gao, Y. (2021). A survey on federated learning. Knowledge-Based Systems, 216:106775.

Zhang, Z., Zhu, H., and Xie, M. (2024). Differential privacy may have a potential optimization effect on some swarm intelligence algorithms besides privacy-preserving. Information Sciences, 654:119870.