TrajSyn: Privacy-Preserving Dataset Distillation from Federated Model Trajectories for Server-Side Adversarial Training
This work addresses the challenge of applying adversarial training in FL settings with strict privacy constraints, offering a solution for safety-critical applications on edge devices.
The paper tackled the problem of adversarial vulnerability in Federated Learning (FL) models by proposing TrajSyn, a framework that synthesizes a proxy dataset from client model updates to enable server-side adversarial training without accessing raw data, resulting in improved adversarial robustness on image classification benchmarks with no extra client compute burden.
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.