arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1705.07751stat.ML
收藏

基于参数交换的异步分布式大规模学习框架

An Asynchronous Distributed Framework for Large-scale Learning Based on Parameter Exchanges

Bikash Joshi, Franck Iutzeler, Massih-Reza Amini

摘要:

在许多分布式学习问题中,计算机器的异构负载可能会损害同步策略的整体性能。在本文中,我们提出了一个有效的异步分布式框架的平滑函数的总和最小化,其中每台机器在其本地函数上并行执行迭代,并异步更新共享参数。通过这种方式,所有机器都可以连续工作,即使它们没有最新版本的共享参数。我们证明了梯度迭代的一般分布式异步方法的一致性的收敛性,然后显示其效率的矩阵分解问题的推荐系统和二进制分类。

英文摘要:

In many distributed learning problems, the heterogeneous loading of computing machines may harm the overall performance of synchronous strategies. In this paper, we propose an effective asynchronous distributed framework for the minimization of a sum of smooth functions, where each machine performs iterations in parallel on its local function and updates a shared parameter asynchronously. In this way, all machines can continuously work even though they do not have the latest version of the shared parameter. We prove the convergence of the consistency of this general distributed asynchronous method for gradient iterations then show its efficiency on the matrix factorization problem for recommender systems and on binary classification.

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。

↑