On the Byzantine-Resilience of Distillation-Based Federated Learning

Christophe Roux; Max Zimmer; Sebastian Pokutta

ICLR 2025poster3 citations

On the Byzantine-Resilience of Distillation-Based Federated Learning

Christophe Roux, Max Zimmer, Sebastian Pokutta

Abstract

Federated Learning (FL) algorithms using Knowledge Distillation (KD) have received increasing attention due to their favorable properties with respect to privacy, non-i.i.d. data and communication cost. These methods depart from transmitting model parameters and instead communicate information about a learning task by sharing predictions on a public dataset. In this work, we study the performance of such approaches in the byzantine setting, where a subset of the clients act in an adversarial manner aiming to disrupt the learning process. We show that KD-based FL algorithms are remarkably resilient and analyze how byzantine clients can influence the learning process. Based on these insights, we introduce two new byzantine attacks and demonstrate their ability to break existing byzantine-resilient methods. Additionally, we propose a novel defence method which enhances the byzantine resilience of KD-based FL algorithms. Finally, we provide a general framework to obfuscate attacks, making them significantly harder to detect, thereby improving their effectiveness.

We show that Knowledge Distillation-based FL is naturally more resilient to malicious clients than parameter-sharing approachesbut we identify novel attack strategies that can still disrupt training and propose effective defenses against them.

BibTeX

@inproceedings{
roux2025on,
title={On the Byzantine-Resilience of Distillation-Based Federated Learning},
author={Christophe Roux and Max Zimmer and Sebastian Pokutta},
booktitle={The Thirteenth International Conference on Learning Representations},
year={2025},
url={https://openreview.net/forum?id=of6EuHT7de}
}