Skip navigation
Please use this identifier to cite or link to this item:
Title: Differential evolution using a localized Cauchy mutation operator
Authors: Thangraj R.
Pant M.
Abraham A.
Deep, K.
Snasel V.
Published in: Proceedings of IEEE International Conference on Systems, Man and Cybernetics
Abstract: In the present work, we propose a new variant of basic DE algorithm called CMDE-G which uses Cauchy mutation (CM) operator. In this algorithm, at the end of every generation, CM is applied as a local search mechanism to explore the neighborhood of the best individual in the population. The performance of CMDE-G algorithm is analyzed on a set of 10 standard benchmark problems and four nontraditional composite functions. Simulation results show that the proposed algorithm helps in improving the solution quality besides maintaining a good convergence rate. ©2010 IEEE.
Citation: Proceedings of IEEE International Conference on Systems, Man and Cybernetics, (2010), 3710- 3716. Istanbul
Issue Date: 2010
Keywords: Cauchy distribution
Differential evolution
Local search
Bench-mark problems
Cauchy distribution
Cauchy mutation
Composite functions
Convergence rates
DE algorithms
Differential Evolution
Local search
Simulation result
Solution quality
Convergence of numerical methods
Evolutionary algorithms
Mathematical operators
ISBN: 9.78E+12
ISSN: 1062922X
Author Scopus IDs: 36810610500
Author Affiliations: Thangraj, R., Indian Institute of Technology Roorkee, India
Pant, M., Indian Institute of Technology Roorkee, India
Abraham, A., Machine Intelligence Research Labs (MIR Labs), Scientific Network for Innovation and Research Excellence, United States
Deep, K., Indian Institute of Technology Roorkee, India
Snasel, V., VSB Technical University of Ostrava, Czech Republic
Corresponding Author: Thangraj, R.; Indian Institute of Technology RoorkeeIndia; email:
Appears in Collections:Conference Publications [MA]

Files in This Item:
There are no files associated with this item.
Show full item record

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