BriefGPT.xyz
Jul, 2011
在线镜像下降的普适性
On the Universality of Online Mirror Descent
HTML
PDF
Nathan Srebro, Karthik Sridharan, Ambuj Tewari
TL;DR
对于一般的凸在线学习问题，我们证明镜像下降法总是可以实现（几乎）最优的遗憾保证。
Abstract
We show that for a general class of
convex online learning
problems,
mirror descent
can always achieve a (nearly) optimal
regret guarantee
→