Change search
ReferencesLink to record
Permanent link

Direct link
Conditional steepest descent directions over Cartesian product sets: With application to the Frank-Wolfe method
Linköping University, Department of Mathematics, Optimization . Linköping University, Faculty of Science & Engineering.
2015 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

We derive a technique for scaling the search directions of feasible direction methods when applied to optimization problems over Cartesian product sets. It is proved that when the scaling is included in a convergent feasible direction method, also the new method will be convergent. The scaling technique is applied to the Frank-Wolfe method, the partanized Frank-Wolfe method and a heuristic Frank-Wolfe method. The performance of  these algorithms with and without scaling is evaluated on the stochastic transportation problem. It is found that the scaling technique has the ability to improve the performance of some methods. In particular we observed a huge improvement in the performance of the partanized Frank-Wolfe method, especially when the scaling is used together with an exact line search and when the number of sets in the Cartesian product is large.

Place, publisher, year, edition, pages
2015. , 61 p.
Keyword [en]
Nonlinear optimization, feasible direction methods, the Frank-Wolfe method, scaled direction, stochastic transportation problem
National Category
URN: urn:nbn:se:liu:diva-123730ISRN: LiTH-MAT-EX--2015/11--SEOAI: diva2:892603
Subject / course
Available from: 2016-01-11 Created: 2016-01-11 Last updated: 2016-01-13Bibliographically approved

Open Access in DiVA

Conditional steepest descent directions over Cartesian product sets: With application to the Frank-Wolfe method(685 kB)66 downloads
File information
File name FULLTEXT02.pdfFile size 685 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Högdahl, Johan
By organisation
Optimization Faculty of Science & Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 81 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 395 hits
ReferencesLink to record
Permanent link

Direct link