Expensive Black-Box Model Optimization Via a Gold Rush Policy

Author:

Isaac Benson1,Allaire Douglas1

Affiliation:

1. Department of Mechanical Engineering, Texas A&M University, College Station, TX 77843 e-mail:

Abstract

The optimization of black-box models is a challenging task owing to the lack of analytic gradient information and structural information about the underlying function, and also due often to significant run times. A common approach to tackling such problems is the implementation of Bayesian global optimization techniques. However, these techniques often rely on surrogate modeling strategies that endow the approximation of the underlying expensive function with nonexistent features. Further, these techniques tend to push new queries away from previously queried design points, making it difficult to locate an optimum point that rests near a previous model evaluation. To overcome these issues, we propose a gold rush (GR) policy that relies on purely local information to identify the next best design alternative to query. The method employs a surrogate constructed pointwise, that adds no additional features to the approximation. The result is a policy that performs well in comparison to state of the art Bayesian global optimization methods on several benchmark problems. The policy is also demonstrated on a constrained optimization problem using a penalty method.

Funder

Air Force Office of Scientific Research

Publisher

ASME International

Subject

Computer Graphics and Computer-Aided Design,Computer Science Applications,Mechanical Engineering,Mechanics of Materials

Reference52 articles.

1. Analysis of Multi-Objective Kriging-Based Methods for Constrained Global Optimization;Comput. Optim. Appl.,2016

2. Exploration of Metamodeling Sampling Criteria for Constrained Global Optimization;Eng. Optim.,2002

3. Ghoreishi, S. F., and Allaire, D. L., 2018, “A Fusion-Based Multi-Information Source Optimization Approach Using Knowledge Gradient Policies,” AIAA Paper No. 2018-1159.10.2514/6.2018-1159

4. Parr, J., Holden, C. M., Forrester, A. I., and Keane, A. J., 2010, “Review of Efficient Surrogate Infill Sampling Criteria With Constraint Handling,” Second International Conference on Engineering Optimization, Lisbon, Portugal, Sept. 6–9, pp. 1–10.https://pdfs.semanticscholar.org/df63/8813760971a61daaa91259ff65b817d0dd51.pdf

5. Metamodeling for High Dimensional Simulation-Based Design Problems;ASME J. Mech. Des.,2010

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3