The Fundamental Price of Secure Aggregation in Differentially Private Federated Learning. (arXiv:2203.03761v1 [cs.LG])

We consider the problem of training a $d$ dimensional model with distributed
differential privacy (DP) where secure aggregation (SecAgg) is used to ensure
that the server only sees the noisy sum of $n$ model updates in every training
round. Taking into account the constraints imposed by SecAgg, we characterize
the fundamental communication cost required to obtain the best accuracy
achievable under $\varepsilon$ central DP (i.e. under a fully trusted server
and no communication constraints). Our results show that $\tilde{O}\left(
\min(n^2\varepsilon^2, …

arxiv federated learning learning price