Two-level Multi-surrogate Assisted Optimization method for high dimensional nonlinear problems
详细信息 查看全文
- 作者：Enying Lia ; Hu Wangb ; c ; wanghu@hnu.edu.cn" class="auth_mail" title="E-mail the corresponding author ; Fan Yeb ; c
- 关键词：AI ; Artificial Immune ; EICAM ; EI criterion assisted modeling ; EGO ; Efficient Global Optimization ; EI ; Expected Improvement ; GA ; Genetic Algorithm ; GMDH ; Group Method of Data Handling ; GSS ; Golden Section Sampling ; HDMR ; High Dimensional Model Representation ; LOOCV ; Leave-One-Out Cross Validation ; MAS ; Multi-surrogate Assisted Sampling algorithm ; EImax ; maximum EI ; MLS ; Moving Least Square ; NTS ; Number of Test Samples ; PR ; Polynomial Regression ; PSO ; Particle Swarm Optimization ; RAAE ; Relative Av
- 刊名：Applied Soft Computing
- 出版年：2016
- 出版时间：September 2016
- 年：2016
- 卷：46
- 期：Complete
- 页码：26-36
- 全文大小：1513 K
文摘
- •
A two layer TMAO is suggested for solving high dimensional problems.
- •
The first-level is based on SVR-HDMR and EI assisted sampling strategy.
- •
The first-level is used to decomposition of high dimensional problems.
- •
The second-level is based on multi-surrogate GMDH framework.
- •
The second-level is used to optimize the correlative sub-problems.