arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1709.07772cs.DCcs.LG
收藏

概率同步并行

Probabilistic Synchronous Parallel

Computer Laboratory, University of Cambridge(剑桥大学计算机实验室)

Liang Wang, Ben Catterall, Richard Mortier

摘要:

大多数机器学习和深度神经网络算法依赖于某些迭代算法来优化其效用/成本函数,例如随机梯度下降。在分布式学习中,网络节点必须协同工作以更新模型参数,并且它们如何进行的方式被称为同步并行设计(或屏障控制)。同步并行协议是任何分布式学习框架的基石,其设计直接影响系统的性能和可扩展性。本文提出了一种新的屏障控制技术-概率同步并行(PSP)。该方案通过在系统中引入采样原语,有效地提高了SGD算法的收敛速度和可扩展性。此外,我们还表明,采样原语可以应用于现有的障碍控制机制,以获得完全分布式的基于PSP的同步并行。我们不仅对基于PSP的SGD算法的收敛性进行了深入的理论分析,而且实现了一个功能齐全的分布式学习框架Actor,并在其上进行了深入的评估。

英文摘要:

Most machine learning and deep neural network algorithms rely on certain iterative algorithms to optimise their utility/cost functions, e.g. Stochastic Gradient Descent. In distributed learning, the networked nodes have to work collaboratively to update the model parameters, and the way how they proceed is referred to as synchronous parallel design (or barrier control). Synchronous parallel protocol is the building block of any distributed learning framework, and its design has direct impact on the performance and scalability of the system. In this paper, we propose a new barrier control technique - Probabilistic Synchronous Parallel (PSP). Com- paring to the previous Bulk Synchronous Parallel (BSP), Stale Synchronous Parallel (SSP), and (Asynchronous Parallel) ASP, the proposed solution e ectively improves both the convergence speed and the scalability of the SGD algorithm by introducing a sampling primitive into the system. Moreover, we also show that the sampling primitive can be applied atop of the existing barrier control mechanisms to derive fully distributed PSP-based synchronous parallel. We not only provide a thorough theoretical analysis1 on the convergence of PSP-based SGD algorithm, but also implement a full-featured distributed learning framework called Actor and perform intensive evaluation atop of it.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑