Unifying mirror descent and dual averaging

10/30/2019

∙

by   Anatoli Juditsky, et al.

∙

0

∙

share

We introduce and analyse a new family of algorithms which generalizes and unifies both the mirror descent and the dual averaging algorithms. The unified analysis of the algorithms involves the introduction of a generalized Bregman divergence which utilizes subgradients instead of gradients. Our approach is general enough to encompass classical settings in convex optimization, online learning, and variational inequalities such as saddle-point problems.