paper-with-me

홈 › Papers

Anytime MiniBatch: Exploiting Stragglers in Online Distributed Optimization

2020-06-10 · ICLR 2019 5 · Nuwan Ferdinand, Haider Al-Lawati, Stark C. Draper, Matthew Nokleby

Distributed optimization is vital in solving large-scale machine learning problems. A widely-shared feature of distributed optimization techniques is the requirement that all nodes complete their assigned tasks in each computational epoch before the system can proceed to the next epoch. In such settings, slow nodes, called stragglers, can greatly slow progress. To mitigate the impact of stragglers, we propose an online distributed optimization method called Anytime Minibatch. In this approach, all nodes are given a fixed time to compute the gradients of as many data samples as possible. The result is a variable per-node minibatch size. Workers then get a fixed communication time to average their minibatch gradients via several rounds of consensus, which are then used to update primal variables via dual averaging. Anytime Minibatch prevents stragglers from holding up the system without wasting the work that stragglers can complete. We present a convergence analysis and analyze the wall time performance. Our numerical results show that our approach is up to 1.5 times faster in Amazon EC2 and it is up to five times faster when there is greater variability in compute node performance.

📄 PDF Abstract BibTeX arXiv:2006.05752

Code (0)

등록된 구현이 없습니다.

Tasks

Distributed Optimization

Similar Papers 제목 키워드 기반

Gradient and Projection Free Distributed Online Min-Max Resource Optimization

2021-12-07 · Jingrong Wang, Ben Liang

We consider distributed online min-max resource allocation with a set of parallel agents and a parameter server. Our goal is to minimize the pointwise maximum over a set of time-varying and decreasing cost functions, wit…

Distributed CPU Scheduling Subject to Nonlinear Constraints

2022-08-30 · Mohammadreza Doostmohammadian, Alireza Aghasi, Apostolos I. Rikos, Andreas Grammenos 외

This paper considers a network of collaborating agents for local resource allocation subject to nonlinear model constraints. In many applications, it is required (or desirable) that the solution be anytime feasible in te…

CPUDistributed OptimizationQuantizationScheduling

Gradient Coding: Avoiding Stragglers in Distributed Learning

2017-08-01 · ICML 2017 8 · Rashish Tandon, Qi Lei, Alexandros G. Dimakis, Nikos Karampatziakis

We propose a novel coding theoretic framework for mitigating stragglers in distributed learning. We show how carefully replicating data blocks and coding across gradients can provide tolerance to failures and stragg…

Nested Gradient Codes for Straggler Mitigation in Distributed Machine Learning

2022-12-16 · Luis Maßny, Christoph Hofmeister, Maximilian Egger, Rawad Bitar 외

We consider distributed learning in the presence of slow and unresponsive worker nodes, referred to as stragglers. In order to mitigate the effect of stragglers, gradient coding redundantly assigns partial computations t…

Scheduling

Distributed Collaborative Hashing and Its Applications in Ant Financial

2018-04-13 · Chaochao Chen, Ziqi Liu, Peilin Zhao, Longfei Li 외

Collaborative filtering, especially latent factor model, has been popularly used in personalized recommendation. Latent factor model aims to learn user and item latent factors from user-item historic behaviors. To apply …

Collaborative Filtering