Abstract
The Max-Cut problem is an NP-hard problem [15]. Extensions of von Neumann’s alternating projections method permit the computation of proximity projections onto convex sets. The present paper exploits this fact by constructing a globally convergent method for the Max-Cut relaxation problem. The feasible set of this relaxed Max-Cut problem is the set of correlation matrices.
| Original language | English |
|---|---|
| Pages (from-to) | 737-746 |
| Number of pages | 10 |
| Journal | Filomat |
| Volume | 31 |
| Issue number | 3 |
| DOIs | |
| State | Published - 2017 |
Bibliographical note
Publisher Copyright:© 2017, University of Nis. All rights reserved.
Keywords
- Alternating projections method
- Global convergence
- Max-Cut problem
ASJC Scopus subject areas
- General Mathematics