Optimization Algorithm Based on Densification and Dynamic Canonical Descent
Bousson, K. B.
Correia, S. D. Correia
Journal of Computational and Applied Mathematics Vol. 191, Nº 2, pp. 269 - 279, July, 2006.
ISSN (print): 0377-0427
Journal Impact Factor: 1,030 (in 0)
Digital Object Identifier: 10.1016/j.cam.2005.07.023
Stochastic methods have gained some popularity in global optimization in that most of them do not assume the cost functions to be differentiable. They have capabilities to avoid being trapped by local optima, and may converge even faster than gradient-based optimization methods on some problems. The present paper proposes an optimization method, which reduces the search space by means of densiﬁcation curves, coupled with the dynamic canonical descent algorithm. The performances of the new method are shown on several known problems classically used for testing optimization algorithms, and proved to outperform competitive algorithms such as simulated annealing and genetic algorithms.