Nesterov-Accelerated Robust Federated Learning Over Byzantine Adversaries
This addresses secure and efficient distributed machine learning for systems vulnerable to malicious attacks, representing an incremental improvement through hybrid methods.
The paper tackles robust federated learning in the presence of Byzantine adversaries by proposing Byrd-NAFL, which integrates Nesterov's momentum with Byzantine-resilient aggregation, achieving faster and safeguarded convergence with validated superiority in speed, accuracy, and resilience over benchmarks.
We investigate robust federated learning, where a group of workers collaboratively train a shared model under the orchestration of a central server in the presence of Byzantine adversaries capable of arbitrary and potentially malicious behaviors. To simultaneously enhance communication efficiency and robustness against such adversaries, we propose a Byzantine-resilient Nesterov-Accelerated Federated Learning (Byrd-NAFL) algorithm. Byrd-NAFL seamlessly integrates Nesterov's momentum into the federated learning process alongside Byzantine-resilient aggregation rules to achieve fast and safeguarding convergence against gradient corruption. We establish a finite-time convergence guarantee for Byrd-NAFL under non-convex and smooth loss functions with relaxed assumption on the aggregated gradients. Extensive numerical experiments validate the effectiveness of Byrd-NAFL and demonstrate the superiority over existing benchmarks in terms of convergence speed, accuracy, and resilience to diverse Byzantine attack strategies.