从客户端模型更新轨迹生成隐私保护数据集，实现无损的服务器端对抗训练。
TrajSyn: Privacy-Preserving Dataset Distillation from Federated Model Trajectories for Server-Side Adversarial Training
- 通过客户端模型更新轨迹合成代理数据集，不访问原始数据。
- 在图像分类任务上显著提升对抗鲁棒性，客户端无需额外计算开销。
- 适合注重隐私与边缘设备算力受限的联邦学习场景。
部署在边缘设备上的深度学习模型在安全关键应用中日益普及。然而，其对对抗扰动的脆弱性带来了重大风险，尤其是在联邦学习（FL）环境中，相同模型分布于数千个客户端。尽管对抗训练是一种强防御手段，但在联邦学习中难以应用，原因在于严格的客户端数据隐私约束以及边缘设备有限的计算能力。本文提出TrajSyn，一个隐私保护框架，通过从客户端模型更新轨迹中合成代理数据集，实现有效的服务器端对抗训练，无需访问原始客户端数据。实验表明，TrajSyn在图像分类基准测试中持续提升对抗鲁棒性，且不增加客户端的计算负担。
原文摘要 · Abstract (English)
Deep learning models deployed on edge devices are increasingly used in safety-critical applications. However, their vulnerability to adversarial perturbations poses significant risks, especially in Federated Learning (FL) settings where identical models are distributed across thousands of clients. While adversarial training is a strong defense, it is difficult to apply in FL due to strict client-data privacy constraints and the limited compute available on edge devices. In this work, we introduce TrajSyn, a privacy-preserving framework that enables effective server-side adversarial training by synthesizing a proxy dataset from the trajectories of client model updates, without accessing raw client data. We show that TrajSyn consistently improves adversarial robustness on image classification benchmarks with no extra compute burden on the client device.
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。