Large-Scale Expensive Optimization with a Switching Strategy
作者机构:School of Computer Science and TechnologyTaiyuan University of Science and TechnologyTaiyuan 030024China Department of Computer Sciences and Information TechnologyUniversity of Kotli Azad Jammu and KashmirKotli 11100Pakistan
出 版 物:《Complex System Modeling and Simulation》 (复杂系统建模与仿真(英文))
年 卷 期:2022年第2卷第3期
页 面:253-263页
核心收录:
学科分类:08[工学] 0812[工学-计算机科学与技术(可授工学、理学学位)]
基 金:This work was supported in part by the National Natural Science Foundation of China(No.61876123) Shanxi Key Research and Development Program(No.202102020101002) Natural Science Foundation of Shanxi Province(Nos.201901D111264 and 201901D111262)
主 题:large-scale optimization problems computationally expensive problems random grouping surrogate models
摘 要:Some optimization problems in scientific research,such as the robustness optimization for the Internet of Things and the neural architecture search,are large-scale in decision space and expensive for objective *** order to get a good solution in a limited budget for the large-scale expensive optimization,a random grouping strategy is adopted to divide the problem into some low-dimensional sub-problems.A surrogate model is then trained for each sub-problem using different strategies to select training data *** that,a dynamic infill criterion is proposed corresponding to the models currently used in the surrogate-assisted sub-problem ***,an escape mechanism is proposed to keep the diversity of the *** performance of the method is evaluated on CEC’2013 benchmark *** results show that the algorithm has better performance in solving expensive large-scale optimization problems.