A study of the neutrality of Boolean function landscapes in genetic programming

Leonardo Vanneschi

doi:10.1016/j.tcs.2011.03.011

A study of the neutrality of Boolean function landscapes in genetic programming

Leonardo Vanneschi

NOVA Information Management School (NOVA IMS)

Research output: Contribution to journal › Article › peer-review

9 Citations (Scopus)

Abstract

The neutrality of genetic programming Boolean function landscapes is investigated in this paper. Compared with some well-known contributions on the same issue, (i) we first define new measures which help in characterizing neutral landscapes; (ii) we use a new sampling methodology, which captures features that are disregarded by uniform random sampling; (iii) we introduce new genetic operators to define the neighborhood of tree structures; and (iv) we compare the fitness landscape induced by different sets of functional operators. This study indicates the existence of a relationship between our neutrality measures and the performance of genetic programming for the problems studied. (C) 2011 Elsevier B.V. All rights reserved.

Original language	Unknown
Pages (from-to)	34-57
Journal	Theoretical Computer Science
Volume	425
Issue number	NA
DOIs	https://doi.org/10.1016/j.tcs.2011.03.011
Publication status	Published - 1 Jan 2012

Access to Document

10.1016/j.tcs.2011.03.011

Cite this

@article{7c772c821e27481d9338afea34a18f8f,

title = "A study of the neutrality of Boolean function landscapes in genetic programming",

abstract = "The neutrality of genetic programming Boolean function landscapes is investigated in this paper. Compared with some well-known contributions on the same issue, (i) we first define new measures which help in characterizing neutral landscapes; (ii) we use a new sampling methodology, which captures features that are disregarded by uniform random sampling; (iii) we introduce new genetic operators to define the neighborhood of tree structures; and (iv) we compare the fitness landscape induced by different sets of functional operators. This study indicates the existence of a relationship between our neutrality measures and the performance of genetic programming for the problems studied. (C) 2011 Elsevier B.V. All rights reserved.",

keywords = "landscapes, crossover, functions, distance, Genetic, Neutrality, slope, difficulty, subtree, Negative, fitness, coefficient, Problem, Boolean, programming, Fitness",

author = "Leonardo Vanneschi",

note = "ISI Document Delivery No.: 920RH Times Cited: 0 Cited Reference Count: 41 Vanneschi, Leonardo Pirola, Yuri Mauri, Giancarlo Tomassini, Marco Collard, Philippe Verel, Sebastien Elsevier science bv Amsterdam",

year = "2012",

month = jan,

day = "1",

doi = "10.1016/j.tcs.2011.03.011",

language = "Unknown",

volume = "425",

pages = "34--57",

journal = "Theoretical Computer Science",

issn = "0304-3975",

publisher = "Elsevier",

number = "NA",

}

TY - JOUR

T1 - A study of the neutrality of Boolean function landscapes in genetic programming

AU - Vanneschi, Leonardo

N1 - ISI Document Delivery No.: 920RH Times Cited: 0 Cited Reference Count: 41 Vanneschi, Leonardo Pirola, Yuri Mauri, Giancarlo Tomassini, Marco Collard, Philippe Verel, Sebastien Elsevier science bv Amsterdam

PY - 2012/1/1

Y1 - 2012/1/1

N2 - The neutrality of genetic programming Boolean function landscapes is investigated in this paper. Compared with some well-known contributions on the same issue, (i) we first define new measures which help in characterizing neutral landscapes; (ii) we use a new sampling methodology, which captures features that are disregarded by uniform random sampling; (iii) we introduce new genetic operators to define the neighborhood of tree structures; and (iv) we compare the fitness landscape induced by different sets of functional operators. This study indicates the existence of a relationship between our neutrality measures and the performance of genetic programming for the problems studied. (C) 2011 Elsevier B.V. All rights reserved.

AB - The neutrality of genetic programming Boolean function landscapes is investigated in this paper. Compared with some well-known contributions on the same issue, (i) we first define new measures which help in characterizing neutral landscapes; (ii) we use a new sampling methodology, which captures features that are disregarded by uniform random sampling; (iii) we introduce new genetic operators to define the neighborhood of tree structures; and (iv) we compare the fitness landscape induced by different sets of functional operators. This study indicates the existence of a relationship between our neutrality measures and the performance of genetic programming for the problems studied. (C) 2011 Elsevier B.V. All rights reserved.

KW - landscapes

KW - crossover

KW - functions

KW - distance

KW - Genetic

KW - Neutrality

KW - slope

KW - difficulty

KW - subtree

KW - Negative

KW - fitness

KW - coefficient

KW - Problem

KW - Boolean

KW - programming

KW - Fitness

U2 - 10.1016/j.tcs.2011.03.011

DO - 10.1016/j.tcs.2011.03.011

M3 - Article

SN - 0304-3975

VL - 425

SP - 34

EP - 57

JO - Theoretical Computer Science

JF - Theoretical Computer Science

IS - NA

ER -