Instinct-based mating in genetic algorithms applied to the tuning of 1-NN classifiers

Thiago Quirino, Miroslav Kubat, Nicholas J. Bryan

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

The behavior of the genetic algorithm (GA), a popular approach to search and optimization problems, is known to depend, among other factors, on the fitness function formula, the recombination operator, and the mutation operator. What has received less attention is the impact of the mating strategy that selects the chromosomes to be paired for recombination. Existing GA implementations mostly choose them probabilistically, according to their fitness function values, but we show that more sophisticated mating strategies can not only accelerate the search, but perhaps even improve the quality of the GA-generated solution. In our implementation, we took inspiration from the "opposites-attract principle that is so common in nature. As a testbed, we chose the problem of 1-NN classifier tuning where genetic solutions have been employed before, and are thus well-understood by the research community. We propose three "instinct-based mating strategies and experimentally investigate their behaviors.

Original languageEnglish
Article number5374400
Pages (from-to)1724-1737
Number of pages14
JournalIEEE Transactions on Knowledge and Data Engineering
Volume22
Issue number12
DOIs
StatePublished - Nov 12 2010

Fingerprint

Classifiers
Tuning
Genetic algorithms
Chromosomes
Testbeds
Mathematical operators

Keywords

  • Genetic algorithm
  • mating strategies
  • multiobjective optimization
  • nearest-neighbor classifiers

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Information Systems
  • Computer Science Applications

Cite this

Instinct-based mating in genetic algorithms applied to the tuning of 1-NN classifiers. / Quirino, Thiago; Kubat, Miroslav; Bryan, Nicholas J.

In: IEEE Transactions on Knowledge and Data Engineering, Vol. 22, No. 12, 5374400, 12.11.2010, p. 1724-1737.

Research output: Contribution to journalArticle

@article{13404fe7d4e548ab8703d8416eebaa30,
title = "Instinct-based mating in genetic algorithms applied to the tuning of 1-NN classifiers",
abstract = "The behavior of the genetic algorithm (GA), a popular approach to search and optimization problems, is known to depend, among other factors, on the fitness function formula, the recombination operator, and the mutation operator. What has received less attention is the impact of the mating strategy that selects the chromosomes to be paired for recombination. Existing GA implementations mostly choose them probabilistically, according to their fitness function values, but we show that more sophisticated mating strategies can not only accelerate the search, but perhaps even improve the quality of the GA-generated solution. In our implementation, we took inspiration from the {"}opposites-attract principle that is so common in nature. As a testbed, we chose the problem of 1-NN classifier tuning where genetic solutions have been employed before, and are thus well-understood by the research community. We propose three {"}instinct-based mating strategies and experimentally investigate their behaviors.",
keywords = "Genetic algorithm, mating strategies, multiobjective optimization, nearest-neighbor classifiers",
author = "Thiago Quirino and Miroslav Kubat and Bryan, {Nicholas J.}",
year = "2010",
month = "11",
day = "12",
doi = "10.1109/TKDE.2009.211",
language = "English",
volume = "22",
pages = "1724--1737",
journal = "IEEE Transactions on Knowledge and Data Engineering",
issn = "1041-4347",
publisher = "IEEE Computer Society",
number = "12",

}

TY - JOUR

T1 - Instinct-based mating in genetic algorithms applied to the tuning of 1-NN classifiers

AU - Quirino, Thiago

AU - Kubat, Miroslav

AU - Bryan, Nicholas J.

PY - 2010/11/12

Y1 - 2010/11/12

N2 - The behavior of the genetic algorithm (GA), a popular approach to search and optimization problems, is known to depend, among other factors, on the fitness function formula, the recombination operator, and the mutation operator. What has received less attention is the impact of the mating strategy that selects the chromosomes to be paired for recombination. Existing GA implementations mostly choose them probabilistically, according to their fitness function values, but we show that more sophisticated mating strategies can not only accelerate the search, but perhaps even improve the quality of the GA-generated solution. In our implementation, we took inspiration from the "opposites-attract principle that is so common in nature. As a testbed, we chose the problem of 1-NN classifier tuning where genetic solutions have been employed before, and are thus well-understood by the research community. We propose three "instinct-based mating strategies and experimentally investigate their behaviors.

AB - The behavior of the genetic algorithm (GA), a popular approach to search and optimization problems, is known to depend, among other factors, on the fitness function formula, the recombination operator, and the mutation operator. What has received less attention is the impact of the mating strategy that selects the chromosomes to be paired for recombination. Existing GA implementations mostly choose them probabilistically, according to their fitness function values, but we show that more sophisticated mating strategies can not only accelerate the search, but perhaps even improve the quality of the GA-generated solution. In our implementation, we took inspiration from the "opposites-attract principle that is so common in nature. As a testbed, we chose the problem of 1-NN classifier tuning where genetic solutions have been employed before, and are thus well-understood by the research community. We propose three "instinct-based mating strategies and experimentally investigate their behaviors.

KW - Genetic algorithm

KW - mating strategies

KW - multiobjective optimization

KW - nearest-neighbor classifiers

UR - http://www.scopus.com/inward/record.url?scp=78149233463&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78149233463&partnerID=8YFLogxK

U2 - 10.1109/TKDE.2009.211

DO - 10.1109/TKDE.2009.211

M3 - Article

AN - SCOPUS:78149233463

VL - 22

SP - 1724

EP - 1737

JO - IEEE Transactions on Knowledge and Data Engineering

JF - IEEE Transactions on Knowledge and Data Engineering

SN - 1041-4347

IS - 12

M1 - 5374400

ER -