A Hybrid Algorithm for Convex Semidefinite Optimization Sören Laue

We present a hybrid algorithm for optimiz-ing a convex, smooth function over the cone of positive semidefinite matrices. Our algo-rithm converges to the global optimal solu-tion and can be used to solve general large-scale semidefinite programs and hence can be readily applied to a variety of machine learn-ing problems. We show experimental results on three machine learning problems. Our approach outperforms state-of-the-art algo-rithms. 1