Proximal Stochastic Dual Coordinate Ascent

Tong Zhang
Rutgers University-Camden

We introduce a proximal version of dual coordinate ascent method. We demonstrate how the derived algorithmic framework can be used for numerous regularized loss minimization problems, including $\ell_1$ regularization and structured output SVM. The convergence rates we obtain match, and sometimes improve, state-of-the-art results.

Presentation (PDF File)

Back to Structure and Randomness in System Identification and Learning