Bregman Divergence for Stochastic Variance Reduction: Saddle-Point and Adversarial Prediction

Open in new window