Multi-fidelity global optimization using a data-mining strategy for computationally intensive black-box problems
作者：
Highlights：
• MFGO realizes the tradeoff between optimization performance and computational cost.
• MFGO achieves a reasonable balance between local exploitation and global exploration.
• MFGO can discover useful knowledge from the LF model via data-mining strategy.
• MFGO shows superior computation efficiency and robustness on various test cases.
摘要
•MFGO realizes the tradeoff between optimization performance and computational cost.•MFGO achieves a reasonable balance between local exploitation and global exploration.•MFGO can discover useful knowledge from the LF model via data-mining strategy.•MFGO shows superior computation efficiency and robustness on various test cases.
论文关键词：Computationally expensive optimization,Global optimization problems,Multi-fidelity optimization,Data mining,Kriging,Surrogate model
论文评审过程：Received 16 April 2021, Revised 31 May 2021, Accepted 9 June 2021, Available online 12 June 2021, Version of Record 19 June 2021.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.107212
