A metamodel-assisted evolutionary method is frequently used to address computationally difficult issues in engineering and science. The radial basis functions (RBF), which is based on a hyperparameter, is a widely used metamodel. The latter requires a numerical technique, which can be time-consuming in and of itself, and this raises the question of whether it is justified, namely when do changes in the hyperparameter result in significant performance increases. This work addresses this issue by presenting a large number of numerical experiments and analysing them to determine when hyperparameter change is justified.
Author(S) Details
Yoel Tenne
Ariel University, Israel.
View Book:- https://stm.bookpi.org/NPER-V4/article/view/5104
No comments:
Post a Comment