arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.04429math.OCcs.LG
收藏

超越随机逼近中的最坏情况分析:矩估计改进实例复杂性

Beyond Worst-Case Analysis in Stochastic Approximation: Moment Estimation Improves Instance Complexity

Tsinghua University(清华大学);Amazon(亚马逊);IBM Research(IBM研究院);Massachusetts Institute of Technology(麻省理工学院)

Jingzhao Zhang, Hongzhou Lin, Subhro Das, Suvrit Sra, Ali Jadbabaie

摘要:

研究了随机逼近问题的梯度方法的预言复杂性。虽然在许多情况下,最优算法和严格的下界是已知的,这些最优算法并没有达到最佳性能时,在实践中使用。我们通过关注实例依赖复杂性而不是最坏情况复杂性来解决这一理论与实践的差距。特别是,我们首先总结了已知的实例依赖的复杂性的结果,并将它们分为三个级别。我们确定了不同层次之间的支配关系,并提出了第四个实例依赖的约束,占主导地位的现有的。然后,我们提供了一个充分条件,根据该条件,自适应算法与矩估计可以实现所提出的界限,而无需噪声水平的知识。我们提出的算法及其分析提供了一个理论上的理由,成功的时刻估计,因为它实现了改进的实例复杂度。

英文摘要:

We study oracle complexity of gradient based methods for stochastic approximation problems. Though in many settings optimal algorithms and tight lower bounds are known for such problems, these optimal algorithms do not achieve the best performance when used in practice. We address this theory-practice gap by focusing on instance-dependent complexity instead of worst case complexity. In particular, we first summarize known instance-dependent complexity results and categorize them into three levels. We identify the domination relation between different levels and propose a fourth instance-dependent bound that dominates existing ones. We then provide a sufficient condition according to which an adaptive algorithm with moment estimation can achieve the proposed bound without knowledge of noise levels. Our proposed algorithm and its analysis provide a theoretical justification for the success of moment estimation as it achieves improved instance complexity.

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。

↑