FedADMM: A federated primal-dual algorithm allowing partial participation
FedADMM: A federated primal-dual algorithm allowing partial participation
复制标题
DOI：
10.1109/cdc51059.2022.9992745
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Han Wang;Siddartha Marella;James Anderson
中科院分区：
文献类型：
--
作者：
Han Wang;Siddartha Marella;James Anderson
联合学习是分布式优化的框架，重点是沟通效率。特别是，它遵循客户服务器广播模型，并且由于其能够适应客户计算和存储资源的异质性的能力而引人注目。数据假设和数据隐私。我们的贡献是提供一种新的联邦学习算法Fedadmm，用于解决非平滑正规化器的非凸复合优化问题。当并非所有客户都能够在非常通用的采样模型下参与给定的通信回合时，我们证明了Fedadmm的融合。
Federated learning is a framework for distributed optimization that places emphasis on communication efficiency. In particular, it follows a client-server broadcast model and is appealing because of its ability to accommodate heterogeneity in client compute and storage resources, non-i.i.d. data assumptions, and data privacy. Our contribution is to offer a new federated learning algorithm, FedADMM, for solving non-convex composite optimization problems with non-smooth regularizers. We prove the convergence of FedADMM for the case when not all clients are able to participate in a given communication round under a very general sampling model.