arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1908.02725math.OCcs.LG
收藏

缩小SVRG理论与实践之间的差距

Towards closing the gap between the theory and practice of SVRG

Télécom Paris(巴黎电信学院);IPP(巴黎理工学院);Princeton University(普林斯顿大学);INRIA(法国国家信息与自动化研究所);Ecole Normale Supérieure(巴黎高等师范学院);PSL Research University(巴黎文理研究大学)

Othmane Sebbouh, Nidham Gazagnadou, Samy Jelassi, Francis Bach, Robert M. Gower

摘要:

最早用于解决经验风险最小化问题的方差缩减随机方法之一是SVRG方法(Johnson & Zhang 2013)。SVRG是基于内外循环的方法,其中在外循环中评估参考全梯度,之后执行内循环的$m \in \mathbb{N}$步骤,其中参考梯度用于构建当前梯度的方差减小估计。SVRG方法及其分析的简单性导致了甚至非凸优化的多种扩展和变体。我们提供了一个更一般的分析SVRG比以前做的任意采样,这使我们能够分析几乎所有形式的迷你通过一个单一的定理。此外,我们的分析重点关注SVRG的更实用变体,包括无环SVRG的新变体(Hofman et al 2015,Kovalev et al 2019,Kulunchakov and Mairal 2019)和k-SVRG的变体(Raj and Stich 2018),其中$m=n$和$n$是数据点的数量。由于我们的设置和分析反映了在实践中所做的,我们能够设置的参数,如小批量大小和步长使用我们的理论在这样一种方式,产生一个更有效的算法在实践中,我们在广泛的数值实验中显示。

英文摘要:

Among the very first variance reduced stochastic methods for solving the empirical risk minimization problem was the SVRG method (Johnson & Zhang 2013). SVRG is an inner-outer loop based method, where in the outer loop a reference full gradient is evaluated, after which $m \in \mathbb{N}$ steps of an inner loop are executed where the reference gradient is used to build a variance reduced estimate of the current gradient. The simplicity of the SVRG method and its analysis have led to multiple extensions and variants for even non-convex optimization. We provide a more general analysis of SVRG than had been previously done by using arbitrary sampling, which allows us to analyse virtually all forms of mini-batching through a single theorem. Furthermore, our analysis is focused on more practical variants of SVRG including a new variant of the loopless SVRG (Hofman et al 2015, Kovalev et al 2019, Kulunchakov and Mairal 2019) and a variant of k-SVRG (Raj and Stich 2018) where $m=n$ and where $n$ is the number of data points. Since our setup and analysis reflect what is done in practice, we are able to set the parameters such as the mini-batch size and step size using our theory in such a way that produces a more efficient algorithm in practice, as we show in extensive numerical experiments.

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。

↑