Methods for improving the design and performance of evolutionary algorithms (original) (raw)
AI-generated Abstract
This dissertation explores the enhancement of evolutionary algorithms (EAs) through the adaptation of multivariate quantitative genetics theory. While EAs are celebrated for their modularity and adaptability, customizations often lead to ineffective algorithms without systematic guidance. The research introduces a general equation of population variance dynamics to monitor key characteristics like exploration and exploitation during EA runs, facilitating better insights for diagnosing and resolving issues. The efficacy of this approach is exemplified through case studies on Pittsburgh approach rule systems and genetic programming trees.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.