Open Access Open Access  Restricted Access Subscription or Fee Access

A Comprehensive Study On Optimization Of Neural Networks And Global Minimization

M. Aravindh, C. Velmurugan

Abstract


The neural network toolbox is one of the commonly used, powerful, commercially available software tools for the constructing new networks and forms a inter linkage between each other. The software is user-friendly, permits flexibility and convenience in interfacing with other toolboxes in the same environment to develop a full application the wide range of applications and has a wide range of scope in making models with linear and non-linear functionalities without any assumptions literally and Applied in almost every field of science and engineering. In addition, it underlies in one directional linear functionalities with and without hard limit, and possess the functionalities of radial and triangular biasing, and suitable for competitive soft max functions. A wide variety of training and learning algorithms are supported. The functions used in the neural networks are as same as brain and nervous system model which are highly parallel & possess high speed information resemble the brain than a serial computer .The functionalities in neural networks have very simple principles and posses very complex behaviors. In this paper an attempt is made to enhance easy operation of neural networks for optimizing and to obtain global minimization through MATLAB and step by step procedures for optimization are included for easy accessibility on neural networks.

Keywords – optimization, global minimization, neural networks, training, mat lab

Full Text:

PDF

References


i. Howard Demuth Mark Beale ,2012,’Neural Network tool box’ the use with toolbox ,reference text book.

ii. Ciiare Mandi 2007.’ Introduction to neural networks’ reference, text book

iii. Wlodislaw duch &jerry korczaz 2008 ‘Neural networks & global optimization method suitable for neural networks’, text book

iv. Walter e.lillo1991 Neural network for constrained optimization problem, reference text book

v. Michail G.lagodakis 1996 Neural networks & optimization problem, reference text book

vi. C.Maheswaran ‘Investigation on the wear behavior of aluminum-6061 and aluminum oxide hybrid composites USING ANN’ International journal of current engineering and technology’,vol.1,no.3,pp.363-369.

vii. S. Haykin, 1999. Neural Networks-a Comprehensive Foundation, Prentice Hall

viii. J. J. Hopfield, D. W. Tank, 1985. “Neural” Computation of Decisions in Optimization Problem, Biological Cybernetics, vol.52, pp.141-152.

ix. .N. Ansari, E. Hou, 1997. Computational Intelligence for Optimization, Norwell, MA: Kluwer Academic Publishing Group.

x. Saravana kumar.G & Kalra P.K .’Optimization by neural networks’ reference text book.


Refbacks

  • There are currently no refbacks.