On the Nesterov-Todd direction in semidefinite programming

Michael Todd, Kim Toh, and Reha Tutuncu

Nesterov and Todd discuss several path-following and potential-reduction interior-point methods for certain convex programming problems. In the special case of semidefinite programming, we discuss how to compute the corresponding directions efficiently, how to view them as Newton directions, and how to take Mehrotra predictor-corrector steps in this framework. We also provide some computational results suggesting that our algorithm provides higher accuracy than alternative methods.

Technical Report 1154, School of Operations Research and Industrial Engineering, Cornell University, Ithaca, NY.

Contact: tutuncu@orie.cornell.edu