Training RBF Neural Network by hybrid Adaptive Modified Particle Swarm Optimization  Tabu Search algorithm for Function Approximation

Full Text(PDF, ) PP.13801387


Author(s) 
Mohammad Alipour Varmazabadi, Ali Solyemani Aiouri 

KEYWORDS 
Function Approximation , metaheuristics , neural network , Particle Swarm Optimization , RBF , Tabu Search , gradient
descent 

ABSTRACT 
This days interests in metaheuristic optimization algorithms have been grown and this algorithms have been used in many fields . Because of this , many algorithms have born or mixed up together in order to get into better results . In this paper a new hybrid AMPSOTS ( Adaptive Modified Particle Swarm Optimization  Tabu Search ) method is presented for training RBF (Radial Basis Function ) neural network that used to function approximation . This AMPSOTS method uses the search ability of Tabu Search algorithm to optimize the Adaptive Modified Particle Swarm Optimization to result in more accurate . This hybrid method is used for learning RBF neural network and uses it for function approximation . Both PSO and TS method use the strategy of best solution of neighborhood in every iteration to get into best result , but every one of them has issues . By use of advantages of this methods we proposed the new method that uses TS to optimize the constants of PSO . numerical result show that the proposed method has improved results contrast to the simple PSO and gradient descent method to learn the RBF neural Network . 

References 

[1] Kennedy, J.; Eberhart, R. (1995). Particle Swarm Optimization. Proceedings of IEEE International Conference on Neural Networks. IV. pp. 1942–1948.
[2] Shi, Y.; Eberhart, R.C. (1998). A modified particle swarm optimizer. Proceedings of IEEE International Conference on Evolutionary
Computation. pp. 69–73.
[3] F. Glover and C. McMillan . The general employee scheduling problem: an integration of MS and AI. Computers and Operations Research,1986.
[4] J. J. HOPFIELD Neural networks and physical systems with emergent collective computational abilities. Proc. NatL Acad. Sci. USA
Vol. 79, pp. 25542558, April 1982 Biophysics
[5] R.C. Eberhart and J. Kennedy A New Optimizer using Particle
Swarm Theory. In Proceedings of the Sixth International Symposium
on Micromachine and Human Science, 1995, pp 39–43.
[6] J. Kennedy. Small Worlds and MegaMinds: Effects of Neighborhood
Topology on Particle Swarm Performance. In Proceedings of the
IEEE Congress on Evolutionary Computation, volume 3,July
1999,pages 1931–1938.
[7] J. Kennedy and R. Mendes Population Structure and Particle Performance. In Proceedings of the IEEE Congress on Evolutionary
Computation, 2002. pages 1671– 1676. IEEE Press.
[8] J. Kennedy, R.C. Eberhart, and Y. Shi. Swarm Intelligence. Morgan
Kaufmann, 2001.
[9] W.K. Chen, Linear Networks and Systems. Belmont, Calif.: Wadsworth, pp. 123135, 1993. (Book style)
[10] F.Glover, M.Laguna, Tabu Search, Univ. Boulder, 1997.
[11] A.hertz, E.Taillard, D.De werra, A Tutorial on Tabu Search,
ORSpektrum 11, 1989, 131141.
[12] A.Sebastian and P.Kumar and M.P.Schoen.AStudy on Hybridization
of Particle Swarm and Tabu Search Algorithms for unconstrained
Optimization and estimation Problems, 14th WSEAS international
conference on systems 2010.pp.458463.
[13] M.Pant, R.Thangaraj, A.Abraham. A New PSO Algorithm with Crossover Operator for Global Optimization Problems. SpringerVerlag
Berlin Heidelberg 2007.
[14] N.Alamelumangai, J.D.Shree. PSO Aided Neuro Fuzzy Inference
System for Ultrasound Image Segmentation. International Journal of
Computer Applications, volume 7No.14,October 2010.
[15] A.G.Bros. Introduction of the Radial Basis Function (RBF) Networks.
Online Symposium of Electronics Engineering. issue 1, vol. 1, DSP
Algorithms: Multimedia, Feb. 13 2001, pp. 17.
[16] N.Holden,A.A.Frietas. A Hybrid PSO/ACO Algorithm for classification GECCO (Companion) 2007: 27452750
[17] M.VakilBaghmisheh. N.Pasevic. Training RBF networks with selective back propagation . Elsevier, neurocomputing 62, 2004, pp.3964.
[18] J.Q.Li, Q.K.Pan, S.X.Xie, B.X.Jia, Y.T.Wang. A hybrid particle swarm
optimization and tabu search algorithm for flexible jobshop scheduling problem. International Journal of Computer Theory and Engineering, vol 2, No.2 April, 2010.
[19] Y.Zhang. L.Wu. A Hybrid TSPSO Optimization Algorithms. Journal
of Convergence Information Technology, volume 6, No.5, May 2011.
[20] S.Ferrari, R.Robbert, F.Stengel. Smooth FunctionApproximation Using Neural Network. IEEE transaction on neural network, vol 16,
No.1, Jan.2005.
[21] Z.Zainuddin, O.Pauline. Function Approximation Using Artificial
Neural Network. WSEAS transaction on Mathematics, issue 6, Vol 7,
June 2008.
[22] J.Brownlee. Clever Algorithms: NatureInspired Programming Recipes. First Edition, Lulu Enterprises, January 2011. ISBN: 97814467
85065.
[23] S.L.Ho, S.Yang, G.ni, H.C.Wong. A Particle Swarm Optimization
Method With Enhanced Global Search Ability for Design Optimizations of Electromagnetic Devices. IEEE transaction on magnetic. Vol.
42, No.4, April 2006.


