You are in the accessibility menu

Please use this identifier to cite or link to this item: http://acervodigital.unesp.br/handle/11449/38376
Title: 
Neural approach for solving several types of optimization problems
Author(s): 
Institution: 
  • Universidade Estadual Paulista (UNESP)
  • Universidade Estadual de Campinas (UNICAMP)
  • Fed Ctr Educ Technol
ISSN: 
0022-3239
Abstract: 
Neural networks consist of highly interconnected and parallel nonlinear processing elements that are shown to be extremely effective in computation. This paper presents an architecture of recurrent neural net-works that can be used to solve several classes of optimization problems. More specifically, a modified Hopfield network is developed and its inter-nal parameters are computed explicitly using the valid-subspace technique. These parameters guarantee the convergence of the network to the equilibrium points, which represent a solution of the problem considered. The problems that can be treated by the proposed approach include combinatorial optimiza-tion problems, dynamic programming problems, and nonlinear optimization problems.
Issue Date: 
1-Mar-2006
Citation: 
Journal of Optimization Theory and Applications. New York: Springer/plenum Publishers, v. 128, n. 3, p. 563-580, 2006.
Time Duration: 
563-580
Publisher: 
Springer
Keywords: 
  • recurrent neural networks
  • nonlinear optimization
  • dynamic programming
  • combinatorial optimization
  • Hopfield network
Source: 
http://dx.doi.org/10.1007/s10957-006-9032-9
URI: 
Access Rights: 
Acesso restrito
Type: 
outro
Source:
http://repositorio.unesp.br/handle/11449/38376
Appears in Collections:Artigos, TCCs, Teses e Dissertações da Unesp

There are no files associated with this item.
 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.