An Evolutionary Algorithm Assisted by an Ensemble of Pareto-Optimal Surrogate Models
2026-08-03 • Neural and Evolutionary Computing
Neural and Evolutionary Computing
AI summaryⓘ
The authors developed a new method that combines several simpler prediction models to better estimate solutions in a type of optimization called surrogate-assisted evolutionary algorithms. Their approach automatically adjusts the smoothness of these prediction models by balancing accuracy and complexity, which helps avoid mistakes from overfitting or underfitting. They also created a way for these different models to work together when picking promising solutions. Testing showed their method performed better than existing ones on various benchmark and real-world problems. The authors have made their code publicly available for others to use.
surrogate modelensemble learningsurrogate-assisted evolutionary algorithm (SAEA)radial basis function network (RBFN)fitness landscapeoverfittingunderfittingbi-objective optimizationmodel complexityoptimization benchmark
Authors
Kei Nishihara, Yaochu Jin, Masaya Nakata
Abstract
An ensemble of surrogate models helps improve the prediction quality and robustness of surrogate models, and in turn, the search performance of surrogate-assisted evolutionary algorithms (SAEAs). Although different degrees of smoothness of the approximated fitness landscapes need to be carefully designed for an effective ensemble, little attention has been paid to the explicit tuning of the degree of smoothness derived by surrogate models. This study proposes an adaptive ensemble SAEA, which automatically constructs plausible ensemble models by optimizing their parameter settings. Unlike existing adaptive/ensemble SAEAs, which consider prediction accuracy alone, the proposed algorithm optimizes the structure of radial basis function networks (RBFNs) by solving bi-objective minimization problems of approximation error and model complexity, resulting in robust ensemble models of accurate surrogate models with different degrees of smoothness of the approximated fitness landscapes. As a result, the over/under-fittings are reduced. Additionally, an infill criterion is designed so that surrogate models with different degrees of smoothness can contribute to the solution prescreening. The experimental results demonstrated the statistical superiority of our algorithm over state-of-the-art SAEAs on a single-objective benchmark and real-world problem sets under an expensive optimization scenario. The source code of the proposed algorithm is available at https://github.com/haranychan/EPOS