CCE Theses and Dissertations

Investigating the Role of Sexual Reproduction in Diploidy Age-Structured Evolutionary Populations

Date of Award

2003

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Graduate School of Computer and Information Sciences

Advisor

Gregory Simco

Committee Member

Sumitra Mukherjee

Committee Member

Junping Sun

Abstract

The John Holland genetic algorithm lacks some of the genomic regulatory mechanisms necessary to overcome novel environmental dynamics by fast adaptive change. This simple genetic algorithm uses a haploid chromosome-based cell that reproduces only once during its single generation life span by using a modified asexual reproductive technique. This original cellular structure does not support the expression of recessive genomic characteristics that are essential for adaptive change. Its single life span and reproductive cycle do not allow for the possibility of highly fit parents to continuously contribute to the successful evolution of the population. Asexual reproductive techniques fail to pass the test of natural selection. This research developed a new genetic algorithm that is based on an object-oriented representation of those structural and behavioral characteristics found in highly adaptive organisms. By using a diploid chromosome based cell that sexually reproduce across many generations; the research has produced a more adaptive optimization algorithm than originally produced by Holland when performance tested using De Jong's Test Functions.

This document is currently not available here.

  Link to NovaCat

Share

COinS