Network random keys : a tree network representation scheme for genetic and evolutionary algorithms

Publication type:
Journal article
Metadata:
Autoren
Sammlungen
  • metadata
ISSN
1063-6560
Ausgabe der Veröffentlichung
1
Zeitschrift
Evolutionary computation
Schlüsselwörter
  • 330 Wirtschaft
  • 330 Economics
Sprache
eng
Paginierung
Seiten: 75 - 97
Datum der Veröffentlichung
2002
Datum der Datenerfassung
2020
Datum, an dem der Datensatz öffentlich gemacht wurde
2020
Zugang
Public
Titel
Network random keys : a tree network representation scheme for genetic and evolutionary algorithms
Ausgabe der Zeitschrift
10

Data source: METADATA.UB

Other metadata sources:
Autoren
Autoren-URL
https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=fis-test-1&SrcAuth=WosAPI&KeyUT=WOS:000174052900005&DestLinkType=FullRecord&DestApp=WOS_CPL
DOI
10.1162/106365602317301781
eISSN
1530-9304
Externe Identifier
  • Clarivate Analytics Document Solution ID: 525DF
  • PubMed Identifier: 11911783
ISSN
1063-6560
Ausgabe der Veröffentlichung
1
Zeitschrift
EVOLUTIONARY COMPUTATION
Schlüsselwörter
  • network random keys
  • characteristic vector encoding
  • tree network
  • representation
  • stealth mutation
  • random keys
Paginierung
75 - 97
Datum der Veröffentlichung
2002
Status
Published
Titel
Network random keys - A tree representation scheme for genetic and evolutionary algorithms
Sub types
  • Article
Ausgabe der Zeitschrift
10

Data source: Web of Science (Lite)

Abstract
<jats:p> When using genetic and evolutionary algorithms for network design, choosing a good representation scheme for the construction of the genotype is important for algorithm performance. One of the most common representation schemes for networks is the characteristic vector representation. However, with encoding trees, and using crossover and mutation, invalid individuals occur that are either under or overspecified. When constructing the offspring or repairing the invalid individuals that do not represent a tree, it is impossible to distinguish between the importance of the links that should be used. These problems can be overcome by transferring the concept of random keys from scheduling and ordering problems to the encoding of trees. This paper investigates the performance of a simple genetic algorithm (SGA) using network random keys (NetKeys) for the one-max tree and a real-world problem. The comparison between the network random keys and the characteristic vector encoding shows that despite the effects of stealth mutation, which favors the characteristic vector representation, selectorecombinative SGAs with NetKeys have some advantages for small and easy optimization problems. With more complex problems, SGAs with network random keys significantly outperform SGAs using characteristic vectors. </jats:p><jats:p> This paper shows that random keys can be used for the encoding of trees, and that genetic algorithms using network random keys are able to solve complex tree problems much faster than when using the characteristic vector. Users should therefore be encouraged to use network random keys for the representation of trees. </jats:p>
Autoren
DOI
10.1162/106365602317301781
eISSN
1530-9304
ISSN
1063-6560
Ausgabe der Veröffentlichung
1
Zeitschrift
Evolutionary Computation
Sprache
en
Paginierung
75 - 97
Datum der Veröffentlichung
2002
Status
Published
Herausgeber
MIT Press - Journals
Herausgeber URL
http://dx.doi.org/10.1162/106365602317301781
Datum der Datenerfassung
2021
Titel
Network Random Keys—A Tree Representation Scheme for Genetic and Evolutionary Algorithms
Ausgabe der Zeitschrift
10

Data source: Crossref

Abstract
When using genetic and evolutionary algorithms for network design, choosing a good representation scheme for the construction of the genotype is important for algorithm performance. One of the most common representation schemes for networks is the characteristic vector representation. However, with encoding trees, and using crossover and mutation, invalid individuals occur that are either under- or over-specified. When constructing the offspring or repairing the invalid individuals that do not represent a tree, it is impossible to distinguish between the importance of the links that should be used. These problems can be overcome by transferring the concept of random keys from scheduling and ordering problems to the encoding of trees. This paper investigates the performance of a simple genetic algorithm (SGA) using network random keys (NetKeys) for the one-max tree and a real-world problem. The comparison between the network random keys and the characteristic vector encoding shows that despite the effects of stealth mutation, which favors the characteristic vector representation, selectorecombinative SGAs with NetKeys have some advantages for small and easy optimization problems. With more complex problems, SGAs with network random keys significantly outperform SGAs using characteristic vectors. This paper shows that random keys can be used for the encoding of trees, and that genetic algorithms using network random keys are able to solve complex tree problems much faster than when using the characteristic vector. Users should therefore be encouraged to use network random keys for the representation of trees.
Addresses
  • Department of Information Systems, University of Bayreuth, Universitätsstr. 30, D-95440 Bayreuth, Germany. rothlauf@uni-bayreuth.de
Autoren
DOI
10.1162/106365602317301781
eISSN
1530-9304
Externe Identifier
  • PubMed Identifier: 11911783
Open access
false
ISSN
1063-6560
Ausgabe der Veröffentlichung
1
Zeitschrift
Evolutionary computation
Schlüsselwörter
  • Random Allocation
  • Evolution, Molecular
  • Algorithms
  • Models, Genetic
  • Research Design
  • Biological Evolution
Sprache
eng
Medium
Print
Paginierung
75 - 97
Datum der Veröffentlichung
2002
Status
Published
Datum der Datenerfassung
2002
Titel
Network random keys - a tree representation scheme for genetic and evolutionary algorithms.
Sub types
  • Review
  • Journal Article
Ausgabe der Zeitschrift
10

Data source: Europe PubMed Central

Abstract
When using genetic and evolutionary algorithms for network design, choosing a good representation scheme for the construction of the genotype is important for algorithm performance. One of the most common representation schemes for networks is the characteristic vector representation. However, with encoding trees, and using crossover and mutation, invalid individuals occur that are either under- or over-specified. When constructing the offspring or repairing the invalid individuals that do not represent a tree, it is impossible to distinguish between the importance of the links that should be used. These problems can be overcome by transferring the concept of random keys from scheduling and ordering problems to the encoding of trees. This paper investigates the performance of a simple genetic algorithm (SGA) using network random keys (NetKeys) for the one-max tree and a real-world problem. The comparison between the network random keys and the characteristic vector encoding shows that despite the effects of stealth mutation, which favors the characteristic vector representation, selectorecombinative SGAs with NetKeys have some advantages for small and easy optimization problems. With more complex problems, SGAs with network random keys significantly outperform SGAs using characteristic vectors. This paper shows that random keys can be used for the encoding of trees, and that genetic algorithms using network random keys are able to solve complex tree problems much faster than when using the characteristic vector. Users should therefore be encouraged to use network random keys for the representation of trees.
Autoren
Autoren-URL
https://www.ncbi.nlm.nih.gov/pubmed/11911783
DOI
10.1162/106365602317301781
ISSN
1063-6560
Ausgabe der Veröffentlichung
1
Zeitschrift
Evol Comput
Schlüsselwörter
  • Algorithms
  • Biological Evolution
  • Evolution, Molecular
  • Models, Genetic
  • Random Allocation
  • Research Design
Sprache
eng
Country
United States
Paginierung
75 - 97
Datum der Veröffentlichung
2002
Status
Published
Datum, an dem der Datensatz öffentlich gemacht wurde
2002
Titel
Network random keys - a tree representation scheme for genetic and evolutionary algorithms.
Sub types
  • Journal Article
  • Review
Ausgabe der Zeitschrift
10

Data source: PubMed

Autoren
DOI
10.1162/106365602317301781
Zeitschrift
Evol. Comput.
Artikelnummer
1
Paginierung
75 - 97
Datum der Veröffentlichung
2002
Titel
Network Random Keys-A Tree Representation Scheme for Genetic and Evolutionary Algorithms.
Ausgabe der Zeitschrift
10

Data source: DBLP

Autoren
Ausgabe der Veröffentlichung
1
Zeitschrift
Evol. Comput. Spring
Paginierung
75 - 97
Datum der Veröffentlichung
2002
Titel
Network random keys a tree network representation scheme for genetic and evolutionary algorithms
Ausgabe der Zeitschrift
10

Data source: CiNii EN

Beziehungen:
Property of

Tools