Monday, May 10
MS6
Semidefinite Programming
10:45 AM-12:45 PM
Room: Georgia 1 & 2
Semidefinite Programming (SDP) is an extension of linear programming
in Euclidean space to the space of real symmetric matrices.
Primal-dual interior-point methods developed for linear programs have
been successfully extended to SDPs, and SDPs have attracted much
attention in various fields such as control system engineering,
global optimization, combinatorial optimization, and truss topology
design. The speakers in this minisymposium will discuss theoretical
and algorithmic issues in interior-point methods for SDPs and
semidefinite programming relaxation of nonconvex programs.
Organizer: Masakazu Kojima
Tokyo Institute of Technology, Japan
- 10:45-11:10 Search Directions for Primal-Dual Interior Point
Methods in Semidefinite Programming
- Kim-Chuan Toh, National University of Singapore, Singapore
- 11:15-11:40 Incomplete Orthogonalization Preconditioner for
Solving Large and Dense Linear Systems which Arise from Semidefinite Programming
- Shao-Liang Zhang, University of Tokyo, Japan; Masakazu
Kojima, Organizer; and Kazuhide Nakata, Nihon Sun Micro Systems,
Tokyo, Japan
- 11:45-12:10 Spectral Relations for Symmetric Cones and Their
Use in Interior Point Methods
- Jos F. Sturm, Maastricht University, The Netherlands
- 12:15-12:40 Discretization and Localization in Successive
Convex Relaxation Methods for Nonconvex Quadratic Optimization Problems
- Masakazu Kojima, Organizer; and Levent Tuncel,
University of Waterloo, Canada
MMD, 2/2/99