Variable-size Memory Evolutionary Algorithm: Studies on Replacing Strategies and Diversity in Dynamic Environments

Authors

Abstract

We investigate some improvements to a memory-based evolutionary algorithm already studied with success in dynamic optimization problems. Two new replacing strategies to incorporate in the algorithm are proposed and a comparative study with previous approaches is made. The results show that the studied mechanisms powerfully improve the efficiency and the adaptability of the evolutionary algorithm.

Keywords

Dynamic Environments, Memory, Diversity, Replacing Strategies

Subject

Evolutionary Optimization

Conference

Genetic and Evolutionary Computation Conference (GECCO-2007), July 2007

Cited by

Year 2009 : 1 citations

Richter, H and Yang, S. X.
Learning behaviour in abstract memory schemes for dynamic optimization problems.
Soft Computing, 13 (12): 1163-1173 October 2009