Adaptive explicit decision functions for probabilistic design and optimization using support vector machines

Anirban Basudhar, Samy Missoum

Research output: Contribution to journalArticle

128 Citations (Scopus)

Abstract

This article presents a methodology to generate explicit decision functions using support vector machines (SVM). A decision function is defined as the boundary between two regions of a design space (e.g., an optimization constraint or a limit-state function in reliability). The SVM-based decision function, which is initially constructed based on a design of experiments, depends on the amount and quality of the training data used. For this reason, an adaptive sampling scheme that updates the decision function is proposed. An accurate approximated explicit decision functions is obtained with a reduced number of function evaluations. Three problems are presented to demonstrate the efficiency of the update scheme to explicitly reconstruct known analytical decision functions. The chosen functions are the boundaries of disjoint regions of the design space. A convergence criterion and error measure are proposed. The scheme is also applied to the definition of an explicit failure region boundary in the case of the buckling of a geometrically nonlinear arch.

Original languageEnglish (US)
Pages (from-to)1904-1917
Number of pages14
JournalComputers and Structures
Volume86
Issue number19-20
DOIs
StatePublished - Oct 2008

Fingerprint

Support vector machines
Support Vector Machine
Optimization
Update
Adaptive Sampling
Convergence Criteria
Design
Function evaluation
Design of Experiments
Arch
Arches
Evaluation Function
Buckling
Design of experiments
Disjoint
Sampling
Methodology
Demonstrate

Keywords

  • Discontinuities
  • Disjoint failure regions
  • Explicit decision functions
  • Optimization
  • Probabilistic design
  • Support vector machines

ASJC Scopus subject areas

  • Computer Science Applications
  • Computational Mechanics

Cite this

Adaptive explicit decision functions for probabilistic design and optimization using support vector machines. / Basudhar, Anirban; Missoum, Samy.

In: Computers and Structures, Vol. 86, No. 19-20, 10.2008, p. 1904-1917.

Research output: Contribution to journalArticle

@article{ce8df57533fe44b999a7b9db02b20ca3,
title = "Adaptive explicit decision functions for probabilistic design and optimization using support vector machines",
abstract = "This article presents a methodology to generate explicit decision functions using support vector machines (SVM). A decision function is defined as the boundary between two regions of a design space (e.g., an optimization constraint or a limit-state function in reliability). The SVM-based decision function, which is initially constructed based on a design of experiments, depends on the amount and quality of the training data used. For this reason, an adaptive sampling scheme that updates the decision function is proposed. An accurate approximated explicit decision functions is obtained with a reduced number of function evaluations. Three problems are presented to demonstrate the efficiency of the update scheme to explicitly reconstruct known analytical decision functions. The chosen functions are the boundaries of disjoint regions of the design space. A convergence criterion and error measure are proposed. The scheme is also applied to the definition of an explicit failure region boundary in the case of the buckling of a geometrically nonlinear arch.",
keywords = "Discontinuities, Disjoint failure regions, Explicit decision functions, Optimization, Probabilistic design, Support vector machines",
author = "Anirban Basudhar and Samy Missoum",
year = "2008",
month = "10",
doi = "10.1016/j.compstruc.2008.02.008",
language = "English (US)",
volume = "86",
pages = "1904--1917",
journal = "Computers and Structures",
issn = "0045-7949",
publisher = "Elsevier Limited",
number = "19-20",

}

TY - JOUR

T1 - Adaptive explicit decision functions for probabilistic design and optimization using support vector machines

AU - Basudhar, Anirban

AU - Missoum, Samy

PY - 2008/10

Y1 - 2008/10

N2 - This article presents a methodology to generate explicit decision functions using support vector machines (SVM). A decision function is defined as the boundary between two regions of a design space (e.g., an optimization constraint or a limit-state function in reliability). The SVM-based decision function, which is initially constructed based on a design of experiments, depends on the amount and quality of the training data used. For this reason, an adaptive sampling scheme that updates the decision function is proposed. An accurate approximated explicit decision functions is obtained with a reduced number of function evaluations. Three problems are presented to demonstrate the efficiency of the update scheme to explicitly reconstruct known analytical decision functions. The chosen functions are the boundaries of disjoint regions of the design space. A convergence criterion and error measure are proposed. The scheme is also applied to the definition of an explicit failure region boundary in the case of the buckling of a geometrically nonlinear arch.

AB - This article presents a methodology to generate explicit decision functions using support vector machines (SVM). A decision function is defined as the boundary between two regions of a design space (e.g., an optimization constraint or a limit-state function in reliability). The SVM-based decision function, which is initially constructed based on a design of experiments, depends on the amount and quality of the training data used. For this reason, an adaptive sampling scheme that updates the decision function is proposed. An accurate approximated explicit decision functions is obtained with a reduced number of function evaluations. Three problems are presented to demonstrate the efficiency of the update scheme to explicitly reconstruct known analytical decision functions. The chosen functions are the boundaries of disjoint regions of the design space. A convergence criterion and error measure are proposed. The scheme is also applied to the definition of an explicit failure region boundary in the case of the buckling of a geometrically nonlinear arch.

KW - Discontinuities

KW - Disjoint failure regions

KW - Explicit decision functions

KW - Optimization

KW - Probabilistic design

KW - Support vector machines

UR - http://www.scopus.com/inward/record.url?scp=49049115700&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=49049115700&partnerID=8YFLogxK

U2 - 10.1016/j.compstruc.2008.02.008

DO - 10.1016/j.compstruc.2008.02.008

M3 - Article

AN - SCOPUS:49049115700

VL - 86

SP - 1904

EP - 1917

JO - Computers and Structures

JF - Computers and Structures

SN - 0045-7949

IS - 19-20

ER -