BriefGPT.xyz
Jan, 2017
严格鞍点问题的经验风险最小化的快速速率
Fast Rates for Empirical Risk Minimization of Strict Saddle Problems
HTML
PDF
Alon Gonen, Shai Shalev-Shwartz
TL;DR
本文研究非凸风险最小化的样本复杂性，并探索了在此上下文中最小化此类功能的有效算法，结果表明这些算法不仅具有统计稳定性，而且具有良好的泛化能力。
Abstract
We derive bounds on the
sample complexity
of empirical risk minimization (ERM) in the context of minimizing
non-convex risks
that admit the
stric
→