Copula Estimation of Distribution Algorithm with PMLE

Author

Xiaodong Guo ; Lifang Wang ; Jianchao Zeng ; Xueliang Zhang

Author_Institution

Complex Syst. & Comput. Intell. Lab., Taiyuan Univ. of Sci. & Technol., Taiyuan, China

Volume

2

fYear

2011

fDate

26-28 July 2011

Firstpage

1077

Lastpage

1081

Abstract

Estimation of Distribution Algorithms (EDAs) have the task to estimate the distribution model of samples. Copula Estimation of Distribution Algorithms (cEDAs) introduce the copula theory into EDAs which divide the multivariate distribution estimation into two parts: the marginal distribution estimation and the estimation of the dependant structure of variables. The parameter of copula influences the shape of dependant structure. PMLE is used in cEDA to estimate the parameter of copula. The experimental results show that the proposed algorithms are feasible and effective.

Keywords

genetic algorithms; maximum likelihood estimation; parameter estimation; PMLE; cEDA; copula estimation of distribution algorithm; copula parameter estimation; copula theory; genetic algorithms; marginal distribution estimation; maximum likelihood estimation; multivariate distribution estimation; variable dependant structure estimation; Algorithm design and analysis; Evolutionary computation; Joints; MIMICs; Maximum likelihood estimation; Optimization; Estimation of Distribution Algorithms (EDAs); Maximum Likelihood Estimation (MLE); copula;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation (ICNC), 2011 Seventh International Conference on

Conference_Location

Shanghai

ISSN

2157-9555

Print_ISBN

978-1-4244-9950-2

Type

conf

DOI

10.1109/ICNC.2011.6022139

Filename

6022139