Search

Federated Learning (FL) allows multiple data owners to build high-quality deep learning models collaboratively, by sharing only model updates and keeping data on their premises. Even though FL offers privacy-by-design, it is vulnerable to membership inference attacks (MIA), where an adversary tries to determine whether a sample was included in the training data. Existing defenses against MIA cannot offer meaningful privacy protection without significantly hampering the model's utility and causing a non-negligible training overhead. In this paper we analyze the underlying causes of the differences in the model behavior for member and non-member samples, which arise from model overfitting and facilitate MIAs. Accordingly, we propose MemberShield, a generalization-based defense method for MIAs that consists of: (i) one-time preprocessing of each client's training data labels that transforms one-hot encoded labels to soft labels and eventually exploits them in local training, and (ii) early stopping the training when the local model's validation accuracy does not improve on that of the global model for a number of epochs. Extensive empirical evaluations on three widely used datasets and four model architectures demonstrate that MemberShield outperforms state-of-the-art defense methods by delivering substantially better practical privacy protection against all forms of MIAs, while better preserving the target model utility. On top of that, our proposal significantly reduces training time and is straightforward to implement, by just tuning a single hyperparameter., Competing Interests: Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Josep Domingo-Ferrer and Zouhair Haddi reports financial support was provided by European Commission. Josep Domingo-Ferrer and David Sanchez reports financial support was provided by Ministry of Science Technology and Innovations. Josep Domingo-Ferrer and David Sanchez reports financial support was provided by Cybersecurity National Institute. Josep Domingo-Ferrer and David Sanchez reports financial support was provided by Government of Catalonia Agency for Administration of University and Research Grants. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.)