A restarting approach on symmetric rank one update for unconstrained optimization
Two basic disadvantages of the symmetric rank one (SR1) update are that the SR1 update may not preserve positive definiteness when starting with a positive definite approximation and the SR1 update can be undefined. A simple remedy to these problems is to restart the update with the initial approxim...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
Springer
2009
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/12810/ http://link.springer.com/article/10.1007%2Fs10589-007-9115-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| id |
my.upm.eprints.12810 |
|---|---|
| record_format |
eprints |
| spelling |
my.upm.eprints.128102016-01-19T01:39:42Z http://psasir.upm.edu.my/id/eprint/12810/ A restarting approach on symmetric rank one update for unconstrained optimization Leong, Wah June Abu Hassan, Malik Two basic disadvantages of the symmetric rank one (SR1) update are that the SR1 update may not preserve positive definiteness when starting with a positive definite approximation and the SR1 update can be undefined. A simple remedy to these problems is to restart the update with the initial approximation, mostly the identity matrix, whenever these difficulties arise. However, numerical experience shows that restart with the identity matrix is not a good choice. Instead of using the identity matrix we used a positive multiple of the identity matrix. The used positive scaling factor is the optimal solution of the measure defined by the problem—maximize the determinant of the update subject to a bound of one on the largest eigenvalue. This measure is motivated by considering the volume of the symmetric difference of the two ellipsoids, which arise from the current and updated quadratic models in quasi- Newton methods. A replacement in the form of a positive multiple of the identity matrix is provided for the SR1 update when it is not positive definite or undefined. Our experiments indicate that with such simple initial scaling the possibility of an undefined update or the loss of positive definiteness for the SR1 method is avoided on all iterations. Springer 2009 Article PeerReviewed Leong, Wah June and Abu Hassan, Malik (2009) A restarting approach on symmetric rank one update for unconstrained optimization. Computational Optimization and Applications, 42 (3). pp. 327-334. ISSN 0926-6003; ESSN: 1573-2894 http://link.springer.com/article/10.1007%2Fs10589-007-9115-z 10.1007/s10589-007-9115-z |
| institution |
Universiti Putra Malaysia |
| building |
UPM Library |
| collection |
Institutional Repository |
| continent |
Asia |
| country |
Malaysia |
| content_provider |
Universiti Putra Malaysia |
| content_source |
UPM Institutional Repository |
| url_provider |
http://psasir.upm.edu.my/ |
| description |
Two basic disadvantages of the symmetric rank one (SR1) update are that the SR1 update may not preserve positive definiteness when starting with a positive definite approximation and the SR1 update can be undefined. A simple remedy to these problems is to restart the update with the initial approximation, mostly the identity matrix, whenever these difficulties arise. However, numerical experience shows that restart with the identity matrix is not a good choice. Instead of using the identity matrix we used a positive multiple of the identity matrix. The used positive scaling factor is the optimal solution of the measure defined by the problem—maximize the determinant of the update subject to a bound of one on the largest eigenvalue. This measure is motivated by considering the volume of the symmetric difference of the two ellipsoids, which arise from the current and updated quadratic models in quasi-
Newton methods. A replacement in the form of a positive multiple of the identity matrix is provided for the SR1 update when it is not positive definite or undefined.
Our experiments indicate that with such simple initial scaling the possibility of an undefined update or the loss of positive definiteness for the SR1 method is avoided
on all iterations. |
| format |
Article |
| author |
Leong, Wah June Abu Hassan, Malik |
| spellingShingle |
Leong, Wah June Abu Hassan, Malik A restarting approach on symmetric rank one update for unconstrained optimization |
| author_facet |
Leong, Wah June Abu Hassan, Malik |
| author_sort |
Leong, Wah June |
| title |
A restarting approach on symmetric rank one update for unconstrained optimization |
| title_short |
A restarting approach on symmetric rank one update for unconstrained optimization |
| title_full |
A restarting approach on symmetric rank one update for unconstrained optimization |
| title_fullStr |
A restarting approach on symmetric rank one update for unconstrained optimization |
| title_full_unstemmed |
A restarting approach on symmetric rank one update for unconstrained optimization |
| title_sort |
restarting approach on symmetric rank one update for unconstrained optimization |
| publisher |
Springer |
| publishDate |
2009 |
| url |
http://psasir.upm.edu.my/id/eprint/12810/ http://link.springer.com/article/10.1007%2Fs10589-007-9115-z |
| _version_ |
1643825141454471168 |
| score |
13.211869 |