A Survey on Sustainable Surrogate-Based Optimisation

Surrogate-based optimisation (SBO) algorithms are a powerful technique that combine machine learning and to solve expensive problems. This type of problem appears when dealing with computationally simulators or algorithms. By approximating the part surrogate, number function evaluations can be reduced. paper defines sustainable SBO, which consists three aspects: applying SBO application, reducing evaluations, considering computational effort parts SBO. The reviews applications have successfully applied over past years, analyses used framework, surrogate used, aspects, open questions. leads recommendations for researchers working on sustainability-related who want apply as well researchers. It is argued transparency computation resources in developing techniques deal large variables objectives, lead more