联邦学习与元学习中的收敛与准确性权衡

Mar, 2021

联邦学习与元学习中的收敛与准确性权衡

Convergence and Accuracy Trade-Offs in Federated Learning and Meta-Learning

Zachary Charles, Jakub Konečný

TL;DR研究一种称为“本地更新方法”的算法族，它可以泛化许多联邦和元学习算法，并证明对于二次模型，本地更新方法等价于对我们精确表征的代理损失进行一阶优化。

Abstract

We study a family of algorithms, which we refer to as local update methods, generalizing many federated and meta-learning algorithms. We prove that for quadratic models, →