Current browse context:
cs.LG
Change to browse by:
References & Citations
Computer Science > Machine Learning
Title: Adaptive Sampling Distributed Stochastic Variance Reduced Gradient for Heterogeneous Distributed Datasets
(Submitted on 20 Feb 2020 (v1), last revised 17 Nov 2020 (this version, v3))
Abstract: We study distributed optimization algorithms for minimizing the average of \emph{heterogeneous} functions distributed across several machines with a focus on communication efficiency. In such settings, naively using the classical stochastic gradient descent (SGD) or its variants (e.g., SVRG) with a uniform sampling of machines typically yields poor performance. It often leads to the dependence of convergence rate on maximum Lipschitz constant of gradients across the devices. In this paper, we propose a novel \emph{adaptive} sampling of machines specially catered to these settings. Our method relies on an adaptive estimate of local Lipschitz constants base on the information of past gradients. We show that the new way improves the dependence of convergence rate from maximum Lipschitz constant to \emph{average} Lipschitz constant across machines, thereby, significantly accelerating the convergence. Our experiments demonstrate that our method indeed speeds up the convergence of the standard SVRG algorithm in heterogeneous environments.
Submission history
From: Ilqar Ramazanli [view email][v1] Thu, 20 Feb 2020 01:55:52 GMT (811kb,D)
[v2] Fri, 30 Oct 2020 20:01:23 GMT (304kb,D)
[v3] Tue, 17 Nov 2020 05:00:16 GMT (304kb,D)
Link back to: arXiv, form interface, contact.