Year: 2021
Type: conference-paper
Abstract: The efficient global optimization (EGO) algorithm based on the Kriging surrogate model is popular. One of the main problems is to determine the infill sampling criteria. An adaptive distance function is proposed and applied to the EI, PI, and MP criteria. The criteria base on a fix distance is also investigated. Seven test problems were used to evaluate these criteria. The results show that the MMP criterion has poor global search ability. PEI, KB and DMP criteria perform better and have better convergence speed and stability.
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W3207389136
