On the Desired Behaviors of Self-Adaptive Evolutionary Algorithms

Authors:
Hans-Georg Beyer;Kalyanmoy Deb
Affiliations:
-;-
Venue:
PPSN VI Proceedings of the 6th International Conference on Parallel Problem Solving from Nature
Year:
2000

Citing 3
Cited 1

Numerical Optimization of Computer Models

Numerical Optimization of Computer Models
Fuzzy Recombination for the Breeder Genetic Algorithm

Proceedings of the 6th International Conference on Genetic Algorithms
An overview of evolutionary algorithms for parameter optimization

Evolutionary Computation

Basic principles for understanding evolutionary algorithms

Fundamenta Informaticae

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we postulate some desired behaviors of any evolutionary algorithm (EA) to demonstrate self-adapdve properties. Thereafter, by calculating population mean and variance growth equations, we find bounds on parameter values in a number of EA operators which will qualify them to demonstrate the self-adaptive behavior. Further, we show that if the population growth rates of different EAs are similar, similar performance is expected. This allows us to connect different self-adaptive EAs on an identical platform. This may lead us to find a more unified understanding of the working of different EAs.