Friday, July 14

Interior Point Methods for Semidefinite and Nonlinear Programming

9:15 AM-11:15 AM
Rio Mar 5

Sponsored by SIAM Activity Group on Optimization

In the last ten years, the impact of interior methods has extended far beyond linear programming, to semidefinite programming, convex conic programming, and nonconvex nonlinear programming. For semidefinite programming, and more generally, convex conic programming, interior point methods are clearly the method of choice; indeed, they have no serious competitor (except the simplex method, which is applicable only to linear programs). For nonconvex nonlinear programming, interior point methods are starting to challenge the established SQP (successive quadratic programming) methods, but they still have a long way to go.The speakers will present a mix of theoretical and numerical results in these exciting and active research areas.

Organizer: Michael L. Overton
Courant Institute of Mathematical Sciences, New York University, USA
9:15-9:40 Sensitivity in Semidefinite Optimization via Interior-Point Methods
Michael J. Todd, and E. Alper Yildirim, Cornell University, USA
9:45-10:10 On the Condition Number of the Linear Equations Arising in Interior-point Methods
Javier Pena, Carnegie Mellon University, USA
10:15-10:40 Feasibility Control in Optimization
Richard Byrd, University of Colorado, Boulder, USA; Marcelo Marazzi, Northwestern University, USA; and Jorgé Nocedal, Northwestern University, USA
10:45-11:10 Ten Top Issues in Interior Methods for Nonlinearly Constrained Optimization
Margaret H. Wright, Bell Laboratories, Lucent Technologies, USA

©2000, Society for Industrial and Applied Mathematics
Designed by Donaghy's Web Consulting
Created 4/12/00; Updated 4/12/00