A test function with full controllability over overlapping: estimation of distribution algorithms

Authors:
Jen-Hao Chang;Chung-Hsiang Hsueh;Hsuan Lee;Tian-Li Yu;Tsung-Yu Ho
Affiliations:
National Taiwan University, Taipei, Taiwan Roc;National Taiwan University, Taipei, Taiwan Roc;National Taiwan University, Taipei, Taiwan Roc;National Taiwan University, Taipei, Taiwan Roc;National Taiwan University, Taipei, Taiwan Roc
Venue:
Proceedings of the 13th annual conference companion on Genetic and evolutionary computation
Year:
2011

Citing 1
Cited 1

A crossover for complex building blocks overlapping

Proceedings of the 8th annual conference on Genetic and evolutionary computation

A test problem with adjustable degrees of overlap and conflict among subproblems

Proceedings of the 14th annual conference on Genetic and evolutionary computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This work proposes a test function to study overlapping. The test function provides full controllability over overlapping. To achieve full controllability, the building block assigning problem is reduced to a bipartite matching problem which allow us to directly assign extent of overlapping to each gene. At the end, an experiment on overlapping shows that to four chosen crossover methods, the problem difficulty increases exponentially with the extent of overlapping.