TY - GEN
T1 - A Relation Surrogate Model for Expensive Multiobjective Continuous and Combinatorial Optimization
AU - Hao, Hao
AU - Zhou, Aimin
N1 - Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2023
Y1 - 2023
N2 - Currently, the research on expensive optimization problems mainly focuses on continuous problems and ignores combinatorial problems, which exist in many real-world applications. Since in surrogate model assisted evolution algorithms (SAEAs), the surrogate models from the community of machine learning are usually designed from continuous problems, and they are not suitable from combinatorial problems. For this reason, we propose a convolution relation model for both continuous and combinatorial problems. In the new relation model, a sample representation method of a relation map is proposed in the data preparation, and the convolution neural network is used to learn the relationships between pairs of candidate solutions. The new method is embedded into a basic multiobjective evolutionary algorithm and applied to a set of continuous and combinatorial problems. The experimental results suggest that the relation model with the same settings can solve continuous and combinatorial problems, and it has an advantage in terms of problem scalability.
AB - Currently, the research on expensive optimization problems mainly focuses on continuous problems and ignores combinatorial problems, which exist in many real-world applications. Since in surrogate model assisted evolution algorithms (SAEAs), the surrogate models from the community of machine learning are usually designed from continuous problems, and they are not suitable from combinatorial problems. For this reason, we propose a convolution relation model for both continuous and combinatorial problems. In the new relation model, a sample representation method of a relation map is proposed in the data preparation, and the convolution neural network is used to learn the relationships between pairs of candidate solutions. The new method is embedded into a basic multiobjective evolutionary algorithm and applied to a set of continuous and combinatorial problems. The experimental results suggest that the relation model with the same settings can solve continuous and combinatorial problems, and it has an advantage in terms of problem scalability.
KW - Combinatorial problems
KW - Expensive optimization
KW - Multiobjective problem
KW - Relation model
UR - https://www.scopus.com/pages/publications/85151062650
U2 - 10.1007/978-3-031-27250-9_15
DO - 10.1007/978-3-031-27250-9_15
M3 - 会议稿件
AN - SCOPUS:85151062650
SN - 9783031272493
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 205
EP - 217
BT - Evolutionary Multi-Criterion Optimization - 12th International Conference, EMO 2023, Proceedings
A2 - Emmerich, Michael
A2 - Deutz, André
A2 - Wang, Hao
A2 - Kononova, Anna V.
A2 - Naujoks, Boris
A2 - Li, Ke
A2 - Miettinen, Kaisa
A2 - Yevseyeva, Iryna
PB - Springer Science and Business Media Deutschland GmbH
T2 - 12th International Conference on Evolutionary Multi-Criterion Optimization, EMO 2023
Y2 - 20 March 2023 through 24 March 2023
ER -