Implementable VU-decomposition methods for composite optimization
| Full text | |
| Author(s): |
Haeser, G.
Total Authors: 1
|
| Document type: | Journal article |
| Source: | OPERATIONS RESEARCH LETTERS; v. 46, n. 3, p. 295-299, MAY 2018. |
| Web of Science Citations: | 1 |
| Abstract | |
In second-order algorithms, we investigate the relevance of the constant rank of the full set of active constraints in ensuring global convergence to a second-order stationary point under a constraint qualification. We show that second-order stationarity is not expected in the non-constant rank case if the growth of so-called tangent AKKT2 sequences is not controlled. Since no algorithm controls their growth, we argue that there is a theoretical limitation of algorithms in finding second-order stationary points without constant rank assumptions. (C) 2018 Elsevier B.V. All rights reserved. (AU) | |
| FAPESP's process: | 13/05475-7 - Computational methods in optimization |
| Grantee: | Sandra Augusta Santos |
| Support type: | Research Projects - Thematic Grants |
| FAPESP's process: | 16/02092-8 - On the second-order information in nonlinear optimization |
| Grantee: | Gabriel Haeser |
| Support type: | Scholarships abroad - Research |