A Second-Order Saddle Point Method for Time-Varying Optimization | IEEE Conference Publication | IEEE Xplore