arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1406.1102math.NAcs.LGcs.NAstat.COstat.ML
收藏

方差减少随机梯度算法在无强凸性下的线性收敛性

Linear Convergence of Variance-Reduced Stochastic Gradient without Strong Convexity

University of Michigan(密歇根大学)

Pinghua Gong, Jieping Ye

AI总结:

本文提出Prox-SVRG和VRPSG算法,证明在无强凸性条件下,这些算法在约束和正则化问题中实现线性收敛,引入Semi-Strongly Convex不等式作为关键理论贡献。

摘要:

随机梯度算法通过仅使用一个或几个样本估计梯度,具有低的每迭代计算成本。它们在大规模优化问题中被广泛应用。然而,由于梯度计算中的固有方差,随机梯度算法通常收敛缓慢,具有亚线性收敛速率。为加速收敛,一些方差减少随机梯度算法,如近端随机方差减少梯度(Prox-SVRG)算法,最近被提出以解决强凸问题。在强凸条件下,这些方差减少随机梯度算法实现线性收敛速率。然而,许多机器学习问题是凸但非强凸的。在本文中,我们引入Prox-SVRG及其投影变种称为方差减少投影随机梯度(VRPSG)算法,以解决广泛用于机器学习的一类非强凸优化问题。作为本文的主要技术贡献,我们证明了VRPSG和Prox-SVRG在无强凸性条件下实现线性收敛速率。证明中的关键成分是一个半强凸(SSC)不等式,这是首次严格证明用于一类非强凸问题的约束和正则化设置中的不等式。此外,SSC不等式与算法无关,可能用于分析其他随机梯度算法,这可能具有独立价值。据我们所知,这是首次在无强凸性条件下建立方差减少随机梯度算法在解决约束和正则化问题中的线性收敛速率的工作。

英文摘要:

Stochastic gradient algorithms estimate the gradient based on only one or a few samples and enjoy low computational cost per iteration. They have been widely used in large-scale optimization problems. However, stochastic gradient algorithms are usually slow to converge and achieve sub-linear convergence rates, due to the inherent variance in the gradient computation. To accelerate the convergence, some variance-reduced stochastic gradient algorithms, e.g., proximal stochastic variance-reduced gradient (Prox-SVRG) algorithm, have recently been proposed to solve strongly convex problems. Under the strongly convex condition, these variance-reduced stochastic gradient algorithms achieve a linear convergence rate. However, many machine learning problems are convex but not strongly convex. In this paper, we introduce Prox-SVRG and its projected variant called Variance-Reduced Projected Stochastic Gradient (VRPSG) to solve a class of non-strongly convex optimization problems widely used in machine learning. As the main technical contribution of this paper, we show that both VRPSG and Prox-SVRG achieve a linear convergence rate without strong convexity. A key ingredient in our proof is a Semi-Strongly Convex (SSC) inequality which is the first to be rigorously proved for a class of non-strongly convex problems in both constrained and regularized settings. Moreover, the SSC inequality is independent of algorithms and may be applied to analyze other stochastic gradient algorithms besides VRPSG and Prox-SVRG, which may be of independent interest. To the best of our knowledge, this is the first work that establishes the linear convergence rate for the variance-reduced stochastic gradient algorithms on solving both constrained and regularized problems without strong convexity.

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。

↑