Low-Cost Privacy-Aware Decentralized Learning

arXiv:2403.11795v1 Announce Type: new
Abstract: This paper introduces ZIP-DL, a novel privacy-aware decentralized learning (DL) algorithm that relies on adding correlated noise to each model update during the model training process. This technique ensures that the added noise almost neutralizes itself during the aggregation process due to its correlation, thus minimizing the impact on model accuracy. In addition, ZIP-DL does not require multiple communication rounds for noise cancellation, addressing the common trade-off between privacy protection and communication overhead. We provide …

abstract accuracy aggregation algorithm arxiv correlation cost cs.dc cs.lg decentralized impact low model accuracy noise novel paper privacy process training type update zip