On the Universality of Online Mirror Descent
Machine Learning
2011-07-21 v1
Abstract
We show that for a general class of convex online learning problems, Mirror Descent can always achieve a (nearly) optimal regret guarantee.
Cite
@article{arxiv.1107.4080,
title = {On the Universality of Online Mirror Descent},
author = {Nathan Srebro and Karthik Sridharan and Ambuj Tewari},
journal= {arXiv preprint arXiv:1107.4080},
year = {2011}
}