Federated Learning (FL) is a distributed machine learning paradigm that enables multi-client collaborative training without data leaving local devices, thereby effectively protecting data privacy. However, when faced with challenges such as non-independent and identically distribution (Non-IID) data and external attacks (e.g., gradient leakage attacks and Byzantine attacks), traditional FL methods often encounter convergence difficulty and performance degradation. To alleviate these issues, this paper proposes a decentralized and personalized FL method called PDD-FANs, which employs a robust aggregation mechanism based on generative adversarial networks (GANs) to enhance the model's robustness when training with Non-IID data and under malicious attack. To validate the performance of PDD-FANs in Non-IID setting, we performed comparative experiments on two benchmark datasets against baseline methods, and we evaluated the ability of PDD-FANs to resist Byzantine attack through an ablation study. The experimental results show that PDD-FANs not only perform better in Non-IID setting but also better maintain the learning capability when subjected to malicious attack.
Wang, X, Nguyen, LT, Chen, OY, Van Schependom, J, Denissen, S & Nagels, G 2026, PDD-FANs: Personalized Decentralized Federated Adversarial Networks with Defense Mechanism in Non-IID Setting. in 2025 IEEE 24th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom). 2025 IEEE 24th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom), IEEE, pp. 1317-1326. https://doi.org/10.1109/Trustcom66490.2025.00152
Wang, X., Nguyen, L. T., Chen, O. Y., Van Schependom, J., Denissen, S., & Nagels, G. (2026). PDD-FANs: Personalized Decentralized Federated Adversarial Networks with Defense Mechanism in Non-IID Setting. In 2025 IEEE 24th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom) (pp. 1317-1326). (2025 IEEE 24th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)). IEEE. https://doi.org/10.1109/Trustcom66490.2025.00152
@inproceedings{fa6252d4a950435a8d6b6529f2e68cb7,
title = "PDD-FANs: Personalized Decentralized Federated Adversarial Networks with Defense Mechanism in Non-IID Setting",
abstract = "Federated Learning (FL) is a distributed machine learning paradigm that enables multi-client collaborative training without data leaving local devices, thereby effectively protecting data privacy. However, when faced with challenges such as non-independent and identically distribution (Non-IID) data and external attacks (e.g., gradient leakage attacks and Byzantine attacks), traditional FL methods often encounter convergence difficulty and performance degradation. To alleviate these issues, this paper proposes a decentralized and personalized FL method called PDD-FANs, which employs a robust aggregation mechanism based on generative adversarial networks (GANs) to enhance the model's robustness when training with Non-IID data and under malicious attack. To validate the performance of PDD-FANs in Non-IID setting, we performed comparative experiments on two benchmark datasets against baseline methods, and we evaluated the ability of PDD-FANs to resist Byzantine attack through an ablation study. The experimental results show that PDD-FANs not only perform better in Non-IID setting but also better maintain the learning capability when subjected to malicious attack.",
author = "Xinguang Wang and Nguyen, \{Linh Trung\} and Chen, \{Oliver Y\} and \{Van Schependom\}, Jeroen and Stijn Denissen and Guy Nagels",
note = "Publisher Copyright: {\textcopyright} 2025 IEEE.",
year = "2026",
month = feb,
day = "2",
doi = "10.1109/Trustcom66490.2025.00152",
language = "English",
isbn = "979-8-3315-6533-6",
series = "2025 IEEE 24th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)",
publisher = "IEEE",
pages = "1317--1326",
booktitle = "2025 IEEE 24th International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)",
}