用生成模型合成数据，自动识别并抵御联邦学习中的恶意攻击。
Byzantine-Robust Federated Learning Using Generative Adversarial Networks
- 服务器用条件GAN生成数据验证客户端更新，无需外部数据集。
- 在多个基准数据集上准确识别恶意客户端，且整体模型精度保持稳定。
- 适用于多种攻击方式，可无缝集成到标准联邦学习流程中。
联邦学习（FL）允许在不共享原始数据的情况下跨分布式客户端协作训练模型，但其鲁棒性易受拜占庭行为（如数据和模型投毒）威胁。现有防御方法存在根本局限：鲁棒聚合规则的误差下界随客户端异构性增加而上升；基于检测的方法常依赖启发式假设（如固定恶意客户端数量）或需可信外部数据集进行验证。本文提出一种防御框架，通过在服务器端部署条件生成对抗网络（cGAN），合成代表性数据以验证客户端更新。该方法无需外部数据集，能适应多样攻击策略，并可无缝集成至标准联邦学习流程。大量实验表明，该框架在多个基准数据集上能准确区分恶意与良性客户端，同时维持高模型精度。此外，还评估了合成数据的代表性、cGAN训练的计算开销，以及方法的透明性与可扩展性。
原文摘要 · Abstract (English)
Federated learning (FL) enables collaborative model training across distributed clients without sharing raw data, but its robustness is threatened by Byzantine behaviors such as data and model poisoning. Existing defenses face fundamental limitations: robust aggregation rules incur error lower bounds that grow with client heterogeneity, while detection-based methods often rely on heuristics (e.g., a fixed number of malicious clients) or require trusted external datasets for validation. We present a defense framework that addresses these challenges by leveraging a conditional generative adversarial network (cGAN) at the server to synthesize representative data for validating client updates. This approach eliminates reliance on external datasets, adapts to diverse attack strategies, and integrates seamlessly into standard FL workflows. Extensive experiments on benchmark datasets demonstrate that our framework accurately distinguishes malicious from benign clients while maintaining overall model accuracy. Beyond Byzantine robustness, we also examine the representativeness of synthesized data, computational costs of cGAN training, and the transparency and scalability of our approach.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。