Design preference elicitation, derivative-free optimization and support vector machine search

Yi Ren, Panos Y. Papalambros

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

3 Citations (Scopus)

Abstract

In design preference elicitation, we seek to find individuals' design preferences, usually through an interactive process that would need only a very small number of interactions. Such a process is akin to an optimization algorithm that operates with point values of an unknown function and converges in a small number of iterations. In this paper, we assume the existence of individual preference functions and show that the elicitation task can be translated into a derivative-free optimization (DFO) problem. Different from commonly-studied DFO formulations, we restrict the outputs to binary classes discriminating sample points with higher function values from those with lower values, to capture people's natural way of expressing preferences through comparisons. To this end, we propose a heuristic search algorithm using support vector machines (SVM) that can locate near-optimal solutions with a limited number of iterations and a small sampling size. Early experiments with test functions show reliable performance when the function is not noisy. Further, SVM search appears promising in design preference elicitation when the dimensionality of the design variable domain is relatively high.

Original languageEnglish (US)
Title of host publicationProceedings of the ASME Design Engineering Technical Conference
Pages335-343
Number of pages9
Volume1
EditionPARTS A AND B
DOIs
StatePublished - 2010
Externally publishedYes
EventASME 2010 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2010 - Montreal, QC, Canada
Duration: Aug 15 2010Aug 18 2010

Other

OtherASME 2010 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2010
CountryCanada
CityMontreal, QC
Period8/15/108/18/10

Fingerprint

Derivative-free Optimization
Elicitation
Support vector machines
Support Vector Machine
Derivatives
Iteration
Sample point
Heuristic Search
Test function
Value Function
Heuristic algorithm
Search Algorithm
Dimensionality
Optimization Algorithm
Optimal Solution
Binary
Optimization Problem
Converge
Unknown
Design

ASJC Scopus subject areas

  • Mechanical Engineering
  • Computer Graphics and Computer-Aided Design
  • Computer Science Applications
  • Modeling and Simulation

Cite this

Ren, Y., & Papalambros, P. Y. (2010). Design preference elicitation, derivative-free optimization and support vector machine search. In Proceedings of the ASME Design Engineering Technical Conference (PARTS A AND B ed., Vol. 1, pp. 335-343) https://doi.org/10.1115/DETC2010-28475

Design preference elicitation, derivative-free optimization and support vector machine search. / Ren, Yi; Papalambros, Panos Y.

Proceedings of the ASME Design Engineering Technical Conference. Vol. 1 PARTS A AND B. ed. 2010. p. 335-343.

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

Ren, Y & Papalambros, PY 2010, Design preference elicitation, derivative-free optimization and support vector machine search. in Proceedings of the ASME Design Engineering Technical Conference. PARTS A AND B edn, vol. 1, pp. 335-343, ASME 2010 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2010, Montreal, QC, Canada, 8/15/10. https://doi.org/10.1115/DETC2010-28475
Ren Y, Papalambros PY. Design preference elicitation, derivative-free optimization and support vector machine search. In Proceedings of the ASME Design Engineering Technical Conference. PARTS A AND B ed. Vol. 1. 2010. p. 335-343 https://doi.org/10.1115/DETC2010-28475
Ren, Yi ; Papalambros, Panos Y. / Design preference elicitation, derivative-free optimization and support vector machine search. Proceedings of the ASME Design Engineering Technical Conference. Vol. 1 PARTS A AND B. ed. 2010. pp. 335-343
@inproceedings{d8bbacf5b7ec4409bebe89b44f20b570,
title = "Design preference elicitation, derivative-free optimization and support vector machine search",
abstract = "In design preference elicitation, we seek to find individuals' design preferences, usually through an interactive process that would need only a very small number of interactions. Such a process is akin to an optimization algorithm that operates with point values of an unknown function and converges in a small number of iterations. In this paper, we assume the existence of individual preference functions and show that the elicitation task can be translated into a derivative-free optimization (DFO) problem. Different from commonly-studied DFO formulations, we restrict the outputs to binary classes discriminating sample points with higher function values from those with lower values, to capture people's natural way of expressing preferences through comparisons. To this end, we propose a heuristic search algorithm using support vector machines (SVM) that can locate near-optimal solutions with a limited number of iterations and a small sampling size. Early experiments with test functions show reliable performance when the function is not noisy. Further, SVM search appears promising in design preference elicitation when the dimensionality of the design variable domain is relatively high.",
author = "Yi Ren and Papalambros, {Panos Y.}",
year = "2010",
doi = "10.1115/DETC2010-28475",
language = "English (US)",
isbn = "9780791844090",
volume = "1",
pages = "335--343",
booktitle = "Proceedings of the ASME Design Engineering Technical Conference",
edition = "PARTS A AND B",

}

TY - GEN

T1 - Design preference elicitation, derivative-free optimization and support vector machine search

AU - Ren, Yi

AU - Papalambros, Panos Y.

PY - 2010

Y1 - 2010

N2 - In design preference elicitation, we seek to find individuals' design preferences, usually through an interactive process that would need only a very small number of interactions. Such a process is akin to an optimization algorithm that operates with point values of an unknown function and converges in a small number of iterations. In this paper, we assume the existence of individual preference functions and show that the elicitation task can be translated into a derivative-free optimization (DFO) problem. Different from commonly-studied DFO formulations, we restrict the outputs to binary classes discriminating sample points with higher function values from those with lower values, to capture people's natural way of expressing preferences through comparisons. To this end, we propose a heuristic search algorithm using support vector machines (SVM) that can locate near-optimal solutions with a limited number of iterations and a small sampling size. Early experiments with test functions show reliable performance when the function is not noisy. Further, SVM search appears promising in design preference elicitation when the dimensionality of the design variable domain is relatively high.

AB - In design preference elicitation, we seek to find individuals' design preferences, usually through an interactive process that would need only a very small number of interactions. Such a process is akin to an optimization algorithm that operates with point values of an unknown function and converges in a small number of iterations. In this paper, we assume the existence of individual preference functions and show that the elicitation task can be translated into a derivative-free optimization (DFO) problem. Different from commonly-studied DFO formulations, we restrict the outputs to binary classes discriminating sample points with higher function values from those with lower values, to capture people's natural way of expressing preferences through comparisons. To this end, we propose a heuristic search algorithm using support vector machines (SVM) that can locate near-optimal solutions with a limited number of iterations and a small sampling size. Early experiments with test functions show reliable performance when the function is not noisy. Further, SVM search appears promising in design preference elicitation when the dimensionality of the design variable domain is relatively high.

UR - http://www.scopus.com/inward/record.url?scp=80054978357&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80054978357&partnerID=8YFLogxK

U2 - 10.1115/DETC2010-28475

DO - 10.1115/DETC2010-28475

M3 - Conference contribution

AN - SCOPUS:80054978357

SN - 9780791844090

VL - 1

SP - 335

EP - 343

BT - Proceedings of the ASME Design Engineering Technical Conference

ER -