SPriFed-OMP: A Differentially Private Federated Learning Algorithm for Sparse Basis Recovery

arXiv:2402.19016v1 Announce Type: cross
Abstract: Sparse basis recovery is a classical and important statistical learning problem when the number of model dimensions $p$ is much larger than the number of samples $n$. However, there has been little work that studies sparse basis recovery in the Federated Learning (FL) setting, where the client data's differential privacy (DP) must also be simultaneously protected. In particular, the performance guarantees of existing DP-FL algorithms (such as DP-SGD) will degrade significantly when $p \gg n$, …

algorithm arxiv cs.cr cs.lg federated federated learning important private problem recovery studies work