An integrated framework of hybrid evolutionary computations

Kengo Takano, Masafumi Hagiwara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

There are various kinds of evolutionary computations (ECs) and they have their own merits and demerits. For example, PSO (Particle Swarm Optimization) shows high ability during initial period in general, whereas DE (Differential Evolution) shows high ability especially in the latter period in search to find more accurate solutions. This paper proposes a novel and integrated framework to effectively combine the merits of several evolutionary computations. There are five distinctive features in the proposed framework. 1) There are several individual pools, and each pool corresponds to one EC. 2) Parents do not necessarily belong to the same EC: for example, a GA type individual can be a spouse of a PSO type individual. 3) Each incorporated EC hasits own evaluated value (EV), and it changes according to the best fitness value at each generation. 4) The number of individuals in each EC changes according to the EV. 5) All of the individuals have their own lifetime to avoid premature convergence; when an individual meets lifetime, the individual reselect EC, and the probability of each EC to be selected depends on the EV. In the proposed framework, therefore, more individuals are allotted to the ECs which show higher performance than the other at each generation: effective usage of individuals is enabled. In this way, this framework can makeuse of merits of incorporated ECs. Original GA, original PSO and original DE are used to construct a simple proposed framework-based system. We carried out experiments using well-known benchmark functions. The results show that the new system outperformed there incorporated ECs in 9functions out of 13 functions.

Original languageEnglish
Title of host publication2009 IEEE Congress on Evolutionary Computation, CEC 2009
Pages838-845
Number of pages8
DOIs
Publication statusPublished - 2009
Event2009 IEEE Congress on Evolutionary Computation, CEC 2009 - Trondheim, Norway
Duration: 2009 May 182009 May 21

Other

Other2009 IEEE Congress on Evolutionary Computation, CEC 2009
CountryNorway
CityTrondheim
Period09/5/1809/5/21

Fingerprint

Evolutionary Computation
Evolutionary algorithms
Particle swarm optimization (PSO)
Particle Swarm Optimization
Differential Evolution
Lifetime
Framework
Premature Convergence
Fitness
High Performance
Benchmark

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computational Theory and Mathematics
  • Theoretical Computer Science

Cite this

Takano, K., & Hagiwara, M. (2009). An integrated framework of hybrid evolutionary computations. In 2009 IEEE Congress on Evolutionary Computation, CEC 2009 (pp. 838-845). [4983032] https://doi.org/10.1109/CEC.2009.4983032

An integrated framework of hybrid evolutionary computations. / Takano, Kengo; Hagiwara, Masafumi.

2009 IEEE Congress on Evolutionary Computation, CEC 2009. 2009. p. 838-845 4983032.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Takano, K & Hagiwara, M 2009, An integrated framework of hybrid evolutionary computations. in 2009 IEEE Congress on Evolutionary Computation, CEC 2009., 4983032, pp. 838-845, 2009 IEEE Congress on Evolutionary Computation, CEC 2009, Trondheim, Norway, 09/5/18. https://doi.org/10.1109/CEC.2009.4983032
Takano K, Hagiwara M. An integrated framework of hybrid evolutionary computations. In 2009 IEEE Congress on Evolutionary Computation, CEC 2009. 2009. p. 838-845. 4983032 https://doi.org/10.1109/CEC.2009.4983032
Takano, Kengo ; Hagiwara, Masafumi. / An integrated framework of hybrid evolutionary computations. 2009 IEEE Congress on Evolutionary Computation, CEC 2009. 2009. pp. 838-845
@inproceedings{bf7ed7df07e44b32b8d671d1011afc1d,
title = "An integrated framework of hybrid evolutionary computations",
abstract = "There are various kinds of evolutionary computations (ECs) and they have their own merits and demerits. For example, PSO (Particle Swarm Optimization) shows high ability during initial period in general, whereas DE (Differential Evolution) shows high ability especially in the latter period in search to find more accurate solutions. This paper proposes a novel and integrated framework to effectively combine the merits of several evolutionary computations. There are five distinctive features in the proposed framework. 1) There are several individual pools, and each pool corresponds to one EC. 2) Parents do not necessarily belong to the same EC: for example, a GA type individual can be a spouse of a PSO type individual. 3) Each incorporated EC hasits own evaluated value (EV), and it changes according to the best fitness value at each generation. 4) The number of individuals in each EC changes according to the EV. 5) All of the individuals have their own lifetime to avoid premature convergence; when an individual meets lifetime, the individual reselect EC, and the probability of each EC to be selected depends on the EV. In the proposed framework, therefore, more individuals are allotted to the ECs which show higher performance than the other at each generation: effective usage of individuals is enabled. In this way, this framework can makeuse of merits of incorporated ECs. Original GA, original PSO and original DE are used to construct a simple proposed framework-based system. We carried out experiments using well-known benchmark functions. The results show that the new system outperformed there incorporated ECs in 9functions out of 13 functions.",
author = "Kengo Takano and Masafumi Hagiwara",
year = "2009",
doi = "10.1109/CEC.2009.4983032",
language = "English",
isbn = "9781424429592",
pages = "838--845",
booktitle = "2009 IEEE Congress on Evolutionary Computation, CEC 2009",

}

TY - GEN

T1 - An integrated framework of hybrid evolutionary computations

AU - Takano, Kengo

AU - Hagiwara, Masafumi

PY - 2009

Y1 - 2009

N2 - There are various kinds of evolutionary computations (ECs) and they have their own merits and demerits. For example, PSO (Particle Swarm Optimization) shows high ability during initial period in general, whereas DE (Differential Evolution) shows high ability especially in the latter period in search to find more accurate solutions. This paper proposes a novel and integrated framework to effectively combine the merits of several evolutionary computations. There are five distinctive features in the proposed framework. 1) There are several individual pools, and each pool corresponds to one EC. 2) Parents do not necessarily belong to the same EC: for example, a GA type individual can be a spouse of a PSO type individual. 3) Each incorporated EC hasits own evaluated value (EV), and it changes according to the best fitness value at each generation. 4) The number of individuals in each EC changes according to the EV. 5) All of the individuals have their own lifetime to avoid premature convergence; when an individual meets lifetime, the individual reselect EC, and the probability of each EC to be selected depends on the EV. In the proposed framework, therefore, more individuals are allotted to the ECs which show higher performance than the other at each generation: effective usage of individuals is enabled. In this way, this framework can makeuse of merits of incorporated ECs. Original GA, original PSO and original DE are used to construct a simple proposed framework-based system. We carried out experiments using well-known benchmark functions. The results show that the new system outperformed there incorporated ECs in 9functions out of 13 functions.

AB - There are various kinds of evolutionary computations (ECs) and they have their own merits and demerits. For example, PSO (Particle Swarm Optimization) shows high ability during initial period in general, whereas DE (Differential Evolution) shows high ability especially in the latter period in search to find more accurate solutions. This paper proposes a novel and integrated framework to effectively combine the merits of several evolutionary computations. There are five distinctive features in the proposed framework. 1) There are several individual pools, and each pool corresponds to one EC. 2) Parents do not necessarily belong to the same EC: for example, a GA type individual can be a spouse of a PSO type individual. 3) Each incorporated EC hasits own evaluated value (EV), and it changes according to the best fitness value at each generation. 4) The number of individuals in each EC changes according to the EV. 5) All of the individuals have their own lifetime to avoid premature convergence; when an individual meets lifetime, the individual reselect EC, and the probability of each EC to be selected depends on the EV. In the proposed framework, therefore, more individuals are allotted to the ECs which show higher performance than the other at each generation: effective usage of individuals is enabled. In this way, this framework can makeuse of merits of incorporated ECs. Original GA, original PSO and original DE are used to construct a simple proposed framework-based system. We carried out experiments using well-known benchmark functions. The results show that the new system outperformed there incorporated ECs in 9functions out of 13 functions.

UR - http://www.scopus.com/inward/record.url?scp=70449889933&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449889933&partnerID=8YFLogxK

U2 - 10.1109/CEC.2009.4983032

DO - 10.1109/CEC.2009.4983032

M3 - Conference contribution

AN - SCOPUS:70449889933

SN - 9781424429592

SP - 838

EP - 845

BT - 2009 IEEE Congress on Evolutionary Computation, CEC 2009

ER -