Evolutionary Algorithm with Non-parametric Surrogate Model for Tensor Program optimization

2020 ◽  
Author(s):  
Ioannis Gatopoulos ◽  
Romain Lepert ◽  
Auke Wiggers ◽  
Giovanni Mariani ◽  
Jakub Tomczak
Keyword(s):  
Evolutionary Algorithm ◽  
Surrogate Model ◽  
Program Optimization ◽  
Non Parametric

scholarly journals A Parallel Surrogate Model Assisted Evolutionary Algorithm for Electromagnetic Design Optimization

2019 ◽  
Vol 3 (2) ◽  
pp. 93-105 ◽  
Author(s):  
Mobayode O. Akinsolu ◽  
Bo Liu ◽  
Vic Grout ◽  
Pavlos I. Lazaridis ◽  
Maria Evelina Mognaschi ◽  
...  
Keyword(s):  
Evolutionary Algorithm ◽  
Design Optimization ◽  
Surrogate Model ◽  
Electromagnetic Design

Parameter identification of nonlinear rotor-bearing system based on improved kriging surrogate model

2016 ◽  
Vol 23 (5) ◽  
pp. 794-807 ◽  
Author(s):  
Fang Han ◽  
Xinglin Guo ◽  
Changki Mo ◽  
Haiyang Gao ◽  
Peijun Hou
Keyword(s):  
Evolutionary Algorithm ◽  
Surrogate Model ◽  
Computation Time ◽  
Previous Method ◽  
New Method ◽  
Structure Parameters ◽  
Kriging Surrogate Model ◽  
Rotor Bearing ◽  
Robustness To Noise ◽  
Bearing System

This paper presents a new method which can identify the structure parameters (such as the bearing parameters, the nonlinear rub-impact parameters, and so on) of a nonlinear rotor-bearing system. Based on an improved kriging surrogate model and evolutionary algorithm (IKSMEA), the new method can provide more accurate results with less computation time. The initial kriging surrogate model (KSM) is constructed by the samples of varying structure parameters and their response values. According to the identified process, a multi-point addition criterion is proposed and more appropriate predicted points are added to update the KSM. Numerical studies and experimental validation of a nonlinear rotor-bearing system are performed. Comparing to the previous method (KSM and evolutionary algorithm), the new method satisfies the condition of convergence with less updating steps and increased robustness to noise. The identified results indicate that the IKSMEA can identify the nonlinear rotor system more effectively and accurately.

Convergence Analysis of Surrogate Assisted (1+1) Evolutionary Algorithm

2012 ◽  
Vol 263-266 ◽  
pp. 2339-2343
Author(s):  
Ying Ming Jin
Keyword(s):  
Markov Chain ◽  
Evolutionary Algorithm ◽  
Convergence Analysis ◽  
Perturbation Analysis ◽  
Surrogate Model ◽  
Transition Matrix ◽  
Matrix Perturbation ◽  
Finite Markov Chain ◽  
Matrix Perturbation Analysis

This paper analyzes the convergence deviation of surrogate assisted (1+1)EA. A model of surrogate assisted (1+1)EA can be built by the finite markov chain, then we got the transition matrix of this algorithm. The deviation of surrogate model can be expressed by the perturbation of transition matrix. So we can estimate the convergence deviation with the method of matrix perturbation analysis. Analyzing of the convergence changes brought by surrogate model’s deviations can help us to have a better select of the surrogate model.

A Multifractal-Guided Multilevel Surrogate Model-Based Evolutionary Algorithm for Expensive Multiobjective Problems

2017 ◽  
Vol 26 (07) ◽  
pp. 1750109 ◽  
Author(s):  
Dongmei Zhang ◽  
Jianping Liao ◽  
Xiaohui Huang ◽  
Jiaqi Jin
Keyword(s):  
Evolutionary Algorithm ◽  
Surrogate Model ◽  
Optimization Problems ◽  
Standard Test ◽  
Sample Distribution ◽  
Model Based ◽  
Multiobjective Problems ◽  
Global Trend ◽  
Computational Resources ◽  
Model Transition

In applied engineering, there are tremendous optimization problems which are multiobjective problems. Meanwhile, a number of them require large amount of time to evaluate their expensive cost function during optimization procedures. This kind of problems can be either financially expensive due to significant computational resources being required or time expensive due to numerous computational complexity. Aiming to this kind of problems, this paper proposed a multilevel surrogate model-based evolutionary algorithm. The proposed method employs DACE modeling method at the beginning to obtain a global trend in the decision domain. When more and more samples are involved and the sample distribution presents a trend or a manifold, the SVR model is utilized as a second-level surrogate model to achieve a better local search. The model transition is determined by the multifractal analysis on the solution set. Experimental results on ZDT and DTLZ standard test cases demonstrate that the time for EGO modeling can be reduced, and the accuracy can be better balanced by comparing to existing SVR and EGO methods.

Sign in / Sign up

Export Citation Format

Share Document