arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1612.04425math.OCcs.DCcs.NAmath.NAstat.ML
收藏

关于异步并行迭代与无界延迟收敛性分析

On the Convergence of Asynchronous Parallel Iteration with Unbounded Delays

Rensselaer Polytechnic Institute(伦斯勒理工学院);Michigan State University(密歇根州立大学);University of California, Los Angeles(加州大学洛杉矶分校)

Zhimin Peng, Yangyang Xu, Ming Yan, Wotao Yin

AI总结:

本文从概率角度分析了异步并行方法在无界延迟下的收敛性,给出了依赖延迟统计的显式步长公式,并通过实验验证了理论模型的正确性,展示了传统最大延迟诱导步长过于保守的问题。

摘要:

近年来,由于涉及非常大规模数据和大量决策变量的问题,异步并行(async-parallel)迭代算法得到了迅速发展。由于异步性,迭代计算使用的是过时的信息,而延迟是指信息创建后被更新的次数。几乎所有的近期工作都假设最大延迟是有限的,并据此设置步长参数。然而,最大延迟在实践中是未知的。本文从概率角度分析了异步并行方法在无界延迟下的收敛性,给出了依赖延迟统计的显式步长公式。使用p+1个相同处理器,我们实验证实延迟近似服从参数为p的泊松分布,与我们的理论模型相符,因此可以据此设置步长。在凸和非凸优化问题上的模拟验证了我们的分析,并显示现有最大延迟诱导的步长过于保守,通常会减慢算法的收敛速度。

英文摘要:

Recent years have witnessed the surge of asynchronous parallel (async-parallel) iterative algorithms due to problems involving very large-scale data and a large number of decision variables. Because of asynchrony, the iterates are computed with outdated information, and the age of the outdated information, which we call delay, is the number of times it has been updated since its creation. Almost all recent works prove convergence under the assumption of a finite maximum delay and set their stepsize parameters accordingly. However, the maximum delay is practically unknown. This paper presents convergence analysis of an async-parallel method from a probabilistic viewpoint, and it allows for large unbounded delays. An explicit formula of stepsize that guarantees convergence is given depending on delays' statistics. With $p+1$ identical processors, we empirically measured that delays closely follow the Poisson distribution with parameter $p$, matching our theoretical model, and thus the stepsize can be set accordingly. Simulations on both convex and nonconvex optimization problems demonstrate the validness of our analysis and also show that the existing maximum-delay induced stepsize is too conservative, often slowing down the convergence of the algorithm.

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。

↑