Composite Objective Mirror Descent
John Duchi, Shai Shalev-Shwartz, Yoram Singer and Ambuj Tewari
In: COLT 2010(2010).
We present a new method for regularized convex optimization and analyze it under both
online and stochastic optimization settings. In addition to unifying previously known ﬁrstorder algorithms, such as the projected gradient method, mirror descent, and forwardbackward splitting, our method yields new analysis and algorithms. We also derive speciﬁc
instantiations of our method for commonly used regularization functions, such as ℓ1, mixed
norm, and trace-norm.