Variance Reduced Stochastic Proximal Algorithm for AUC Maximization

Stochastic Gradient Descent has been widely studied with classification accuracy as a performance measure. However, these stochastic algorithms are not applicable when non-decomposable pairwise measures used, such Area under the ROC curve (AUC), standard metric used classes imbalanced. Several have proposed for optimizing AUC metric, one of recent being Proximal Algorithm (SPAM). downside gradient descent is that it suffers from high variance leading to very slow convergence. reduced methods faster convergence guarantees than vanilla combat this issue. Again, used. In paper, we develop Variance Reduced algorithm Maximization (VRSPAM) combines two areas analyzing metrics and optimization efforts guarantee We perform an in-depth theoretical empirical analysis demonstrate our converges existing state-of-the-art maximization problem.