Communication Efficient Distributed Newton Method with Fast Convergence Rates

MaRDI QIDQ6438311zbMATH OpenFDO

Authors Chengchang Liu, Lesi Chen, Luo Luo, John C. S. Lui

Publication date 29 May 2023

Has companion code repository https://github.com/7ccliu/c2eden

Abstract: We propose a communication and computation efficient second-order method for distributed optimization. For each iteration, our method only requires $mathcalO(d)$ communication complexity, where $d$ is the problem dimension. We also provide theoretical analysis to show the proposed method has the similar convergence rate as the classical second-order optimization algorithms. Concretely, our method can find~$$-second-order stationary points for nonconvex problem by $$ iterations, where $L$ is the Lipschitz constant of Hessian. Moreover, it enjoys a local superlinear convergence under the strongly-convex assumption. Experiments on both convex and nonconvex problems show that our proposed method performs significantly better than baselines.

This page was built for publication: Communication Efficient Distributed Newton Method with Fast Convergence Rates

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6438311)