arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1808.07576cs.LGcs.DCstat.ML
收藏

Cooperative SGD:通信高效SGD算法设计与分析的统一框架

Cooperative SGD: A unified Framework for the Design and Analysis of Communication-Efficient SGD Algorithms

Carnegie Mellon University(卡内基梅隆大学);Department of Electrical & Computer Engineering, Carnegie Mellon University(卡内基梅隆大学电气与计算机工程系)

Jianyu Wang, Gauri Joshi

摘要:

通信高效的SGD算法允许节点执行本地更新并定期同步本地模型,在提高分布式SGD的速度和可扩展性方面非常有效。然而,一个严格的收敛性分析和不同的通信减少战略的比较研究仍然是一个很大程度上开放的问题。本文提出了一个统一的框架,称为合作SGD,包括现有的通信效率的SGD算法,如平均,弹性平均和分散SGD。通过分析合作SGD,我们提供了新的收敛性保证现有的算法。此外,该框架使我们能够设计新的通信高效的SGD算法,在减少通信开销和实现快速错误收敛与低错误地板之间取得最佳平衡。

英文摘要:

Communication-efficient SGD algorithms, which allow nodes to perform local updates and periodically synchronize local models, are highly effective in improving the speed and scalability of distributed SGD. However, a rigorous convergence analysis and comparative study of different communication-reduction strategies remains a largely open problem. This paper presents a unified framework called Cooperative SGD that subsumes existing communication-efficient SGD algorithms such as periodic-averaging, elastic-averaging and decentralized SGD. By analyzing Cooperative SGD, we provide novel convergence guarantees for existing algorithms. Moreover, this framework enables us to design new communication-efficient SGD algorithms that strike the best balance between reducing communication overhead and achieving fast error convergence with low error floor.

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。

↑