ABSTRACT
Discusses another extension of arc-search interior-point algorithm to the semi-definite programming problem (SDP). Semi-definite programming problem is a relative new area in convex optimization. It has found many new applications and it has efficient interior-point algorithms to solve it. However, introducing arc-search interior-point algorithm for the semi-definite programming problem opens the possibility of finding better interior-point algorithm with improved polynomial bound. As a matter of fact, the arc-search interior-point algorithm proposed in this chapter is proved to have better polynomial bound than its counterparts, the one using traditional line search method. Again, the arc-search interior-point algorithm is a feasible algorithm which needs feasible starting point. For the same reason we stated many times, there is a need to develop infeasible arc-search interior-point algorithm in the future.
