Differentially Private Federated Learning without Noise Addition: When is it Possible?

arXiv:2405.04551v1 Announce Type: cross
Abstract: Federated Learning (FL) with Secure Aggregation (SA) has gained significant attention as a privacy preserving framework for training machine learning models while preventing the server from learning information about users' data from their individual encrypted model updates. Recent research has extended privacy guarantees of FL with SA by bounding the information leakage through the aggregate model over multiple training rounds thanks to leveraging the "noise" from other users' updates. However, the privacy metric used in …

abstract aggregation arxiv attention cs.cr cs.lg data federated learning framework information machine machine learning machine learning models noise privacy privacy preserving research server training type updates while