Difficulty of unimodal and multimodal landscapes in genetic programming

Authors:
Leonardo Vanneschi;Marco Tomassini;Manuel Clergue;Philippe Collard
Affiliations:
Computer Science Institute, University of Lausanne, Lausanne, Switzerland;Computer Science Institute, University of Lausanne, Lausanne, Switzerland;I3S Laboratory, University of Nice, Sophia Antipolis, France;I3S Laboratory, University of Nice, Sophia Antipolis, France
Venue:
GECCO'03 Proceedings of the 2003 international conference on Genetic and evolutionary computation: PartII
Year:
2003

Citing 7
Cited 7

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
Fitness Distance Correlation And Problem Difficulty For Genetic Programming

GECCO '02 Proceedings of the Genetic and Evolutionary Computation Conference
Maintaining the Diversity of Genetic Programs

EuroGP '02 Proceedings of the 5th European Conference on Genetic Programming
An analysis of genetic programming

An analysis of genetic programming
An adverse interaction between crossover and restricted tree depth in genetic programming

GECCO '96 Proceedings of the 1st annual conference on Genetic and evolutionary computation
Fitness distance correlation in structural mutation genetic programming

EuroGP'03 Proceedings of the 6th European conference on Genetic programming

A Study of Fitness Distance Correlation as a Difficulty Measure in Genetic Programming

Evolutionary Computation
The effects of constant neutrality on performance and problem hardness in GP

EuroGP'08 Proceedings of the 11th European conference on Genetic programming
Practical performance models of algorithms in evolutionary program induction and other domains

Artificial Intelligence
A survey of problem difficulty in genetic programming

AI*IA'05 Proceedings of the 9th conference on Advances in Artificial Intelligence
Negative slope coefficient: a measure to characterize genetic programming fitness landscapes

EuroGP'06 Proceedings of the 9th European conference on Genetic Programming
Assessment of problem modality by differential performance of lexicase selection in genetic programming: a preliminary report

Proceedings of the 14th annual conference companion on Genetic and evolutionary computation
Models of performance of evolutionary program induction algorithms based on indicators of problem difficulty

Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an original study of fitness distance correlation as a measure of problem difficulty in genetic programming. A new definition of distance, called structural distance, is used and suitable mutation operators for the program space are defined. The difficulty is studied for a number of problems, including, for the first time in GP, multimodal ones, both for the new hand-tailored mutation operators and standard crossover. Results are in agreement with empirical observations, thus confirming that fitness distance correlation can be considered a reasonable index of difficulty for genetic programming, at least for the set of problems studied here.