BriefGPT.xyz
May, 2012
利用结构加速SGD的随机平滑方法
Stochastic Smoothing for Nonsmooth Minimizations: Accelerating SGD by Exploiting Structure
HTML
PDF
Hua Ouyang, Alexander Gray
TL;DR
本文提出了一种加速的非平滑随机梯度下降算法- ANSGD,该算法利用常见非平滑损失函数的结构来实现一类问题(包括SVM)的最优收敛速率,是第一个能够实现最优O(1/t)率的随机算法来最小化非平滑损失函数的算法,经实证比较表明,ANSGD明显优于以前的次梯度下降算法,包括SGD。
Abstract
In this work we consider the
stochastic minimization
of
nonsmooth convex loss functions
, a central problem in machine learning. We propose a novel algorithm called Accelerated Nonsmooth Stochastic Gradient Descen
→