Distributed stochastic inertial-accelerated methods with delayed derivatives for nonconvex problems

TL;DR

This paper introduces an inertial proximal stochastic subgradient method with delayed derivatives for nonsmooth nonconvex optimization, providing convergence guarantees and demonstrating advantages in distributed asynchronous environments.

Contribution

It proposes a novel inertial accelerated SsGM capable of handling delayed derivatives in distributed settings with proven convergence rates.

Findings

Convergence rate of O(1/K^{1/2}) for three classes of nonconvex problems.

Delayed derivatives slow convergence, but the effect diminishes over iterations.

Higher parallel speed-up observed in asynchronous updates despite delays.

Abstract

Stochastic gradient methods (SGMs) are predominant approaches for solving stochastic optimization. On smooth nonconvex problems, a few acceleration techniques have been applied to improve the convergence rate of SGMs. However, little exploration has been made on applying a certain acceleration technique to a stochastic subgradient method (SsGM) for nonsmooth nonconvex problems. In addition, few efforts have been made to analyze an (accelerated) SsGM with delayed derivatives. The information delay naturally happens in a distributed system, where computing workers do not coordinate with each other. In this paper, we propose an inertial proximal SsGM for solving nonsmooth nonconvex stochastic optimization problems. The proposed method can have guaranteed convergence even with delayed derivative information in a distributed environment. Convergence rate results are established to three classes of nonconvex problems: weakly-convex nonsmooth problems with a convex regularizer, composite nonconvex problems with a nonsmooth convex regularizer, and smooth nonconvex problems. For each problem class, the convergence rate is $O(1/K^{\frac{1}{2}})$ in the expected value of the gradient norm square, for $K$ iterations. In a distributed environment, the convergence rate of the proposed method will be slowed down by the information delay. Nevertheless, the slow-down effect will decay with the number of iterations for the latter two problem classes. We test the proposed method on three applications. The numerical results clearly demonstrate the advantages of using the inertial-based acceleration. Furthermore, we observe higher parallelization speed-up in asynchronous updates over the synchronous counterpart, though the former uses delayed derivatives. Our source code is released at https://github.com/RPI-OPT/Inertial-SsGM