Memory-enhanced univariate marginal distribution algorithms for dynamic optimization problems

Date

2005

Advisors

Journal Title

Journal ISSN

ISSN

DOI

Volume Title

Publisher

IEEE Press

Type

Conference

Peer reviewed

Yes

Abstract

Several approaches have been developed into evolutionary algorithms to deal with dynamic optimization problems, of which memory and random immigrants are two major schemes. This paper investigates the application of a direct memory scheme for univariate marginal distribution algorithms (UMDAs), a class of evolutionary algorithms, for dynamic optimization problems. The interaction between memory and random immigrants for UMDAs in dynamic environments is also investigated. Experimental study shows that the memory scheme is efficient for UMDAs in dynamic environments and that the interactive effect between memory and random immigrants for UMDAs in dynamic environments depends on the dynamic environments.

Description

Keywords

Citation

Yang, S. (2005) Memory-enhanced univariate marginal distribution algorithms for dynamic optimization problems. Proceedings of the 2005 IEEE Congress on Evolutionary Computation, 3, pp. 2560-2567

Rights

Research Institute

Institute of Artificial Intelligence (IAI)