Modification of the Particle Swarm Optimizer for Locating All the Global Minima
In many optimization applications, escaping from the local minima as well as computing all the global minima of an objective function is of vital importance. In this paper the Particle Swarm Optimization method is modified in order to locate and evaluate all the global minima of an objective function. The new approach separates the swarm properly when a candidate minimizer is detected. This technique can also be used for escaping from the local minima which is very important in neural network training.
KeywordsParticle Swarm Optimization Local Search Global Minimizer Inertia Weight Neural Network Training
Unable to display preview. Download preview PDF.
- R.C. Eberhart and Y.H. Shi, “Evolving Artificial Neural Networks”, Proc. Int. Conf. on N.N. and Brain, Beijing, P.R. China, 1998.Google Scholar
- R.C. Eberhart, P.K. Simpson and R.W. Dobbins, “Computational Intelligence PC Tools”, Academic Press Professional, Boston, 1996.Google Scholar
- R. Horst, P.M. Pardalos and N.V. Thoai, “Introduction to Global Optimization”, Kluwer Academic Publishers, 1995.Google Scholar
- J. Kennedy and R.C. Eberhart, “Partide Swarm Optimization”, Proc. IEEE Int. Conf. on N.N., Piscataway, NJ, pp. 1942–1948, 1995.Google Scholar
- K.P. Parsopoulos, V.P. Plagianakos, G.D. Magoulas and M.N. Vrahatis, “Objective function stretching” to alleviate convergence to local minima”, Nonlinear Analysis, T.M.A., 2001, to appear.Google Scholar
- W.H. Press, W.T. Vetterling, S.A. Teukolsky and B.P. Flannery, “Numerical Recipes in Fortran 77”, Cambridge University Press, 1992.Google Scholar