Graduate Student Seminar, 4:30pm February 16, 2004; HH127


Thanasak Mouktonglang


Optimization: Introduction to Interior-point methods, and primal-dual algorithms


Linear programming has been the dominant paradigm in optimization since the development of simplex method in 1940s. In 1984, the paper by Karmarkar started a new area of research for a new class of method known as interior-point method, and in the decade since then primal-dual algorithms have emerged as the most important and useful algorithms from this class. I first talk about a general optimization problems. Some important terminologies will be introduced. Then I'll quickly talk about Linear-programming problems, Second-order cone programming problems, Semidefinite-programming problems and how to solve them. Then if time is permitted, I'll end with a cool application of semidefinite programming problems.

To volunteer to give a talk, or for any other questions regarding this schedule, contact Wesley Calvert