MPI-INF Logo
Campus Event Calendar

Event Entry

What and Who

Computing Integral Solutions for Semidefinite Programs

Lorant Porkolab
MPII
AG1 Mittagsseminar
AG 1  
AG Audience
English

Date, Time and Location

Friday, 12 December 97
13:30
60 Minutes
MPII - 46
024
Saarbrücken

Abstract

Semidefinite programming can be regarded as a generalization of linear

programming in which the positive orthant is replaced by the cone of
symmetric positive semidefinite matrices. In the talk, I will give
an extension of Lenstra's theorem on the polynomial-time solvability
of integer linear programming in fixed dimension to integer semidefinite
programming. In fact, I will consider the more general problem of
computing integral points in arbitrary convex semi-algebraic sets,
present some recent results, and then apply them to the special case of
semidefinite programs to obtain the previously mentioned extension.

(Joint work with Leonid Khachiyan.)

Contact

Lorant Porkolab
(0681) 9325-502
--email hidden
passcode not visible
logged in users only

Tags, Category, Keywords and additional notes

Semi-algebraic Sets; Semidefinite Programming; Algorithmic Complexity