An Empirical Investigation of How Degree Neutrality Affects GP Search

Authors:
Edgar Galván-López;Riccardo Poli
Affiliations:
Natural Computing Research & Applications Group, Complex and Adaptive Systems Lab, University College Dublin,;School of Computer Science and Electonic Engineering, University of Essex, Colchester, UK CO4 3SQ
Venue:
MICAI '09 Proceedings of the 8th Mexican International Conference on Artificial Intelligence
Year:
2009

Citing 11
Cited 1

The royal tree problem, a benchmark for single and multiple population genetic programming

Advances in genetic programming
Fitness Distance Correlation and Ridge Functions

PPSN V Proceedings of the 5th International Conference on Parallel Problem Solving from Nature
Cartesian Genetic Programming

Proceedings of the European Conference on Genetic Programming
Finding Needles in Haystacks Is Not Hard with Neutrality

EuroGP '02 Proceedings of the 5th European Conference on Genetic Programming
Finding needles in haystacks is harder with neutrality

GECCO '05 Proceedings of the 7th annual conference on Genetic and evolutionary computation
A Study of Fitness Distance Correlation as a Difficulty Measure in Genetic Programming

Evolutionary Computation
On the effects of bit-wise neutrality on fitness distance correlation, phenotypic mutation rates and problem hardness

FOGA'07 Proceedings of the 9th international conference on Foundations of genetic algorithms
Fitness distance correlation in structural mutation genetic programming

EuroGP'03 Proceedings of the 6th European conference on Genetic programming
On the limiting distribution of program sizes in tree-based genetic programming

EuroGP'07 Proceedings of the 10th European conference on Genetic programming
The effects of constant neutrality on performance and problem hardness in GP

EuroGP'08 Proceedings of the 11th European conference on Genetic programming
Some steps towards understanding how neutrality affects evolutionary search

PPSN'06 Proceedings of the 9th international conference on Parallel Problem Solving from Nature

Defining locality as a problem difficulty measure in genetic programming

Genetic Programming and Evolvable Machines

Quantified Score

Hi-index	0.00

Visualization

Abstract

Over the last years, neutrality has inspired many researchers in the area of Evolutionary Computation (EC) systems in the hope that it can aid evolution. However, there are contradictory results on the effects of neutrality in evolutionary search. The aim of this paper is to understand how neutrality - named in this paper degree neutrality - affects GP search. For analysis purposes, we use a well-defined measure of hardness (i.e., fitness distance correlation) as an indicator of difficulty in the absence and in the presence of neutrality, we propose a novel approach to normalise distances between a pair of trees and finally, we use a problem with deceptive features where GP is well-known to have poor performance and see the effects of neutrality in GP search.