arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1206.6448cs.LGstat.ML
收藏

在线交替方向法

Online Alternating Direction Method

University of Minnesota, Twin Cities(明尼苏达大学双城校区)

Huahua Wang, Arindam Banerjee

摘要:

在线优化已经成为大规模优化的有力工具。在本文中,我们介绍了有效的在线算法的交替方向法(ADM)的基础上。我们引入了一个新的证明技术,ADM在批处理设置,产生的O(1/T)收敛速度的ADM,形成的基础上遗憾分析在线设置。我们考虑两种情况下,在线设置,根据解决方案是否需要躺在可行集或不。在这两种情况下,我们建立遗憾的目标函数,以及一般和强凸函数的约束违反界。初步的结果来说明所提出的算法的性能。

英文摘要:

Online optimization has emerged as powerful tool in large scale optimization. In this paper, we introduce efficient online algorithms based on the alternating directions method (ADM). We introduce a new proof technique for ADM in the batch setting, which yields the O(1/T) convergence rate of ADM and forms the basis of regret analysis in the online setting. We consider two scenarios in the online setting, based on whether the solution needs to lie in the feasible set or not. In both settings, we establish regret bounds for both the objective function as well as constraint violation for general and strongly convex functions. Preliminary results are presented to illustrate the performance of the proposed algorithms.

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。

↑