On the convergence rate method of an Improved Clonal Selection Algorithm

Author

Hong Lu

Author_Institution

Sch. of Electron. Eng., Huaihai Inst. of Technol., Lianyungang, China

fYear

2011

fDate

22-24 July 2011

Firstpage

5413

Lastpage

5417

Abstract

It is complicated and important to study the covergence rate of clonal selection algorithms in the field of artificial immune computation. However, there are few theoretical studies for it. The classical homogeneous Markov chain analyse is replaced by a new pure probability method, and from the definition of strong convergence in probability, the convergence rate of an Improved Clonal Selection Algorithm(ICSA) is analyzed under some conditions, and a method of estimating the convergence rate of the ICSA is obtained. The simulation results of multi-modal function optimization illustrate the validity of convergence rate method.

Keywords

Markov processes; artificial immune systems; convergence; probability; artificial immune computation; classical homogeneous Markov chain; convergence rate method; improved clonal selection algorithm; multimodal function optimization; probability method; Algorithm design and analysis; Convergence; Electronic mail; Evolutionary computation; Immune system; Markov processes; Optimization; Chaos; Clonal Selection Principle; Convergence; Strong Convergence in Probability;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference (CCC), 2011 30th Chinese

Conference_Location

Yantai

ISSN

1934-1768

Print_ISBN

978-1-4577-0677-6

Electronic_ISBN

1934-1768

Type

conf

Filename

6001085