| Article ID: | iaor20002453 |
| Country: | United States |
| Volume: | 9 |
| Issue: | 1/3 |
| Start Page Number: | 141 |
| End Page Number: | 160 |
| Publication Date: | Jan 1998 |
| Journal: | Optimization Methods & Software |
| Authors: | Nesterov Y. |
In this paper we study the qualitiy of semidefinite relaxation for global quadratic optimization problem with diagonal quadratic constraints. We prove that such relaxation approximates the exact solution of the problem with relative accurancy μ = (π/2) – 1. We consider some applications of this result.