Self-organizing maps by difference of convex functions optimization

We offer an efficient approach based on difference of convex functions (DC) optimization for self-organizing maps (SOM). We consider SOM as an optimization problem with a nonsmooth, nonconvex energy function and investigated DC programming and DC algorithm (DCA), an innovative approach in nonconvex...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Data mining and knowledge discovery Ročník 28; číslo 5-6; s. 1336 - 1365
Hlavní autoři: Le Thi, Hoai An, Nguyen, Manh Cuong
Médium: Journal Article
Jazyk:angličtina
Vydáno: Boston Springer US 01.09.2014
Springer Nature B.V
Springer
Témata:
Algorithms
Artificial Intelligence
Chemistry and Earth Sciences
Computer Science
Control theory
Convex analysis
Data compression
Data mining
Data Mining and Knowledge Discovery
Direct current
Energy
Gene expression
Information Storage and Retrieval
Mapping
Mathematical analysis
Mathematical models
Neighborhoods
Neural networks
Optimization
Optimization techniques
Physics
Programming
Social networks
Statistics for Engineering
Topography
DC programming
Self-organizing maps
DCA
ISSN:1384-5810, 1573-756X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:We offer an efficient approach based on difference of convex functions (DC) optimization for self-organizing maps (SOM). We consider SOM as an optimization problem with a nonsmooth, nonconvex energy function and investigated DC programming and DC algorithm (DCA), an innovative approach in nonconvex optimization framework to effectively solve this problem. Furthermore an appropriate training version of this algorithm is proposed. The numerical results on many real-world datasets show the efficiency of the proposed DCA based algorithms on both quality of solutions and topographic maps.
Bibliografie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-1
ObjectType-Feature-2
content type line 23
ISSN:1384-5810
1573-756X
DOI:10.1007/s10618-014-0369-7