Accuracy vs. robustness: Bi-criteria optimized ensemble of metamodels

Can Cui, Teresa Wu, Mengqi Hu, Jeffery D. Weir, Xianghua Chu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Scopus citations

Abstract

Simulation has been widely used in modeling engineering systems. A metamodel is a surrogate model used to approximate a computationally expensive simulation model. Extensive research has investigated the performance of different metamodeling techniques in terms of accuracy and/or robustness and concluded no model outperforms others across diverse problem structures. Motivated by this finding, this research proposes a bi-criteria (accuracy and robustness) optimized ensemble framework to optimally identify the contributions from each metamodel (Kriging, Support Vector Regression and Radial Basis Function), where uncertainties are modeled for evaluating robustness. Twenty-eight functions from the literature are tested. It is observed for most problems, a Pareto Frontier is obtained, while for some problems only a single point is obtained. Seven geometrical and statistical metrics are introduced to explore the relationships between the function properties and the ensemble models. It is concluded that the bi-criteria optimized ensembles render not only accurate but also robust metamodels.

Original languageEnglish (US)
Title of host publicationProceedings of the 2014 Winter Simulation Conference, WSC 2014
EditorsAndreas Tolk, Levent Yilmaz, Saikou Y. Diallo, Ilya O. Ryzhov
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages616-627
Number of pages12
ISBN (Electronic)9781479974863
DOIs
StatePublished - Jan 23 2015
Event2014 Winter Simulation Conference, WSC 2014 - Savannah, United States
Duration: Dec 7 2014Dec 10 2014

Publication series

NameProceedings - Winter Simulation Conference
Volume2015-January
ISSN (Print)0891-7736

Other

Other2014 Winter Simulation Conference, WSC 2014
CountryUnited States
CitySavannah
Period12/7/1412/10/14

ASJC Scopus subject areas

  • Software
  • Modeling and Simulation
  • Computer Science Applications

Fingerprint Dive into the research topics of 'Accuracy vs. robustness: Bi-criteria optimized ensemble of metamodels'. Together they form a unique fingerprint.

  • Cite this

    Cui, C., Wu, T., Hu, M., Weir, J. D., & Chu, X. (2015). Accuracy vs. robustness: Bi-criteria optimized ensemble of metamodels. In A. Tolk, L. Yilmaz, S. Y. Diallo, & I. O. Ryzhov (Eds.), Proceedings of the 2014 Winter Simulation Conference, WSC 2014 (pp. 616-627). [7019926] (Proceedings - Winter Simulation Conference; Vol. 2015-January). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WSC.2014.7019926