Resumen
Convergence and diversity of solutions play an essential role in the design of multi-objective evolutionary algorithms (MOEAs). Among the available diversity mechanisms, the ϵ-dominance has shown a proper balance between convergence and diversity. When using ϵ-dominance, diversity is ensured by partitioning the objective space into boxes of size ϵ and, typically, a single solution is allowed at each of these boxes. However, there is no easy way to determine the precise value of ϵ. In this paper, we investigate how this goal can be achieved by using a co-evolutionary scheme that looks for the proper values of ϵ along the search without any need of a previous user's knowledge. We include the proposed co-evolutionary scheme into an MOEA based on ϵ-dominance giving rise to a new MOEA. We evaluate the proposed MOEA solving standard benchmark test problems. According to our results, it is a promising alternative for solving multi-objective optimization problems because three main reasons: 1) it is competitive concerning state-of-the-art MOEAs, 2) it does not need extra information about the problem, and 3) it is computationally efficient.
Idioma original | English |
---|---|
Número de artículo | 8637162 |
Páginas (desde-hasta) | 18267-18283 |
Número de páginas | 17 |
Publicación | IEEE Access |
Volumen | 7 |
DOI | |
Estado | Published - 1 ene 2019 |
Huella dactilar
ASJC Scopus subject areas
- Computer Science(all)
- Materials Science(all)
- Engineering(all)
Citar esto
}
A Co-Evolutionary Scheme for Multi-Objective Evolutionary Algorithms Based on ϵ-Dominance. / Menchaca-Mendez, Adriana; Montero, Elizabeth; Antonio, Luis Miguel; Zapotecas-Martinez, Saul; Coello Coello, Carlos A.Coello; Riff, Maria Cristina.
En: IEEE Access, Vol. 7, 8637162, 01.01.2019, p. 18267-18283.Resultado de la investigación: Article
TY - JOUR
T1 - A Co-Evolutionary Scheme for Multi-Objective Evolutionary Algorithms Based on ϵ-Dominance
AU - Menchaca-Mendez, Adriana
AU - Montero, Elizabeth
AU - Antonio, Luis Miguel
AU - Zapotecas-Martinez, Saul
AU - Coello Coello, Carlos A.Coello
AU - Riff, Maria Cristina
PY - 2019/1/1
Y1 - 2019/1/1
N2 - Convergence and diversity of solutions play an essential role in the design of multi-objective evolutionary algorithms (MOEAs). Among the available diversity mechanisms, the ϵ-dominance has shown a proper balance between convergence and diversity. When using ϵ-dominance, diversity is ensured by partitioning the objective space into boxes of size ϵ and, typically, a single solution is allowed at each of these boxes. However, there is no easy way to determine the precise value of ϵ. In this paper, we investigate how this goal can be achieved by using a co-evolutionary scheme that looks for the proper values of ϵ along the search without any need of a previous user's knowledge. We include the proposed co-evolutionary scheme into an MOEA based on ϵ-dominance giving rise to a new MOEA. We evaluate the proposed MOEA solving standard benchmark test problems. According to our results, it is a promising alternative for solving multi-objective optimization problems because three main reasons: 1) it is competitive concerning state-of-the-art MOEAs, 2) it does not need extra information about the problem, and 3) it is computationally efficient.
AB - Convergence and diversity of solutions play an essential role in the design of multi-objective evolutionary algorithms (MOEAs). Among the available diversity mechanisms, the ϵ-dominance has shown a proper balance between convergence and diversity. When using ϵ-dominance, diversity is ensured by partitioning the objective space into boxes of size ϵ and, typically, a single solution is allowed at each of these boxes. However, there is no easy way to determine the precise value of ϵ. In this paper, we investigate how this goal can be achieved by using a co-evolutionary scheme that looks for the proper values of ϵ along the search without any need of a previous user's knowledge. We include the proposed co-evolutionary scheme into an MOEA based on ϵ-dominance giving rise to a new MOEA. We evaluate the proposed MOEA solving standard benchmark test problems. According to our results, it is a promising alternative for solving multi-objective optimization problems because three main reasons: 1) it is competitive concerning state-of-the-art MOEAs, 2) it does not need extra information about the problem, and 3) it is computationally efficient.
KW - co-evolutionary schemes
KW - Multi-objective evolutionary algorithms
KW - parameter setting
KW - ϵ-dominance
UR - http://www.scopus.com/inward/record.url?scp=85062242465&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2019.2896962
DO - 10.1109/ACCESS.2019.2896962
M3 - Article
AN - SCOPUS:85062242465
VL - 7
SP - 18267
EP - 18283
JO - IEEE Access
JF - IEEE Access
SN - 2169-3536
M1 - 8637162
ER -