Design preference elicitation

Exploration and learning

Yi Ren, Panos Papalambros

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

5 Citations (Scopus)

Abstract

We study design preference elicitation, namely discovery of an individual's design preferences, through human-computer interactions. In each interaction, the computer presents a set of designs to the human subject who is then asked to pick preferred designs from the set. The computer learns from this feedback in a cumulative fashion and creates new sets of designs to query the subject. Under the hypothesis that human responses are deterministic, we investigate two interaction algorithms, namely, evolutionary and statistical learning-based, for converging the elicitation process to near-optimally preferred designs. We apply the process to visual preferences for three-dimensional automobile exterior shapes. Evolutionary methods can be useful for design exploration, but learning-based methods have a stronger theoretical foundation and are more successful in eliciting subject preferences efficiently.

Original languageEnglish (US)
Title of host publicationICED 11 - 18th International Conference on Engineering Design - Impacting Society Through Engineering Design
Pages149-158
Number of pages10
Volume10
EditionPART 2
StatePublished - 2011
Externally publishedYes
Event18th International Conference on Engineering Design, ICED 11 - Copenhagen, Denmark
Duration: Aug 15 2011Aug 18 2011

Other

Other18th International Conference on Engineering Design, ICED 11
CountryDenmark
CityCopenhagen
Period8/15/118/18/11

Fingerprint

Human computer interaction
Evolutionary algorithms
Automobiles
Feedback

Keywords

  • Active learning
  • Design preference elicitation
  • Genetic algorithm
  • Statistical learning

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this

Ren, Y., & Papalambros, P. (2011). Design preference elicitation: Exploration and learning. In ICED 11 - 18th International Conference on Engineering Design - Impacting Society Through Engineering Design (PART 2 ed., Vol. 10, pp. 149-158)

Design preference elicitation : Exploration and learning. / Ren, Yi; Papalambros, Panos.

ICED 11 - 18th International Conference on Engineering Design - Impacting Society Through Engineering Design. Vol. 10 PART 2. ed. 2011. p. 149-158.

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

Ren, Y & Papalambros, P 2011, Design preference elicitation: Exploration and learning. in ICED 11 - 18th International Conference on Engineering Design - Impacting Society Through Engineering Design. PART 2 edn, vol. 10, pp. 149-158, 18th International Conference on Engineering Design, ICED 11, Copenhagen, Denmark, 8/15/11.
Ren Y, Papalambros P. Design preference elicitation: Exploration and learning. In ICED 11 - 18th International Conference on Engineering Design - Impacting Society Through Engineering Design. PART 2 ed. Vol. 10. 2011. p. 149-158
Ren, Yi ; Papalambros, Panos. / Design preference elicitation : Exploration and learning. ICED 11 - 18th International Conference on Engineering Design - Impacting Society Through Engineering Design. Vol. 10 PART 2. ed. 2011. pp. 149-158
@inproceedings{32f952b0449f43f281ec71ddb095a924,
title = "Design preference elicitation: Exploration and learning",
abstract = "We study design preference elicitation, namely discovery of an individual's design preferences, through human-computer interactions. In each interaction, the computer presents a set of designs to the human subject who is then asked to pick preferred designs from the set. The computer learns from this feedback in a cumulative fashion and creates new sets of designs to query the subject. Under the hypothesis that human responses are deterministic, we investigate two interaction algorithms, namely, evolutionary and statistical learning-based, for converging the elicitation process to near-optimally preferred designs. We apply the process to visual preferences for three-dimensional automobile exterior shapes. Evolutionary methods can be useful for design exploration, but learning-based methods have a stronger theoretical foundation and are more successful in eliciting subject preferences efficiently.",
keywords = "Active learning, Design preference elicitation, Genetic algorithm, Statistical learning",
author = "Yi Ren and Panos Papalambros",
year = "2011",
language = "English (US)",
isbn = "9781904670308",
volume = "10",
pages = "149--158",
booktitle = "ICED 11 - 18th International Conference on Engineering Design - Impacting Society Through Engineering Design",
edition = "PART 2",

}

TY - GEN

T1 - Design preference elicitation

T2 - Exploration and learning

AU - Ren, Yi

AU - Papalambros, Panos

PY - 2011

Y1 - 2011

N2 - We study design preference elicitation, namely discovery of an individual's design preferences, through human-computer interactions. In each interaction, the computer presents a set of designs to the human subject who is then asked to pick preferred designs from the set. The computer learns from this feedback in a cumulative fashion and creates new sets of designs to query the subject. Under the hypothesis that human responses are deterministic, we investigate two interaction algorithms, namely, evolutionary and statistical learning-based, for converging the elicitation process to near-optimally preferred designs. We apply the process to visual preferences for three-dimensional automobile exterior shapes. Evolutionary methods can be useful for design exploration, but learning-based methods have a stronger theoretical foundation and are more successful in eliciting subject preferences efficiently.

AB - We study design preference elicitation, namely discovery of an individual's design preferences, through human-computer interactions. In each interaction, the computer presents a set of designs to the human subject who is then asked to pick preferred designs from the set. The computer learns from this feedback in a cumulative fashion and creates new sets of designs to query the subject. Under the hypothesis that human responses are deterministic, we investigate two interaction algorithms, namely, evolutionary and statistical learning-based, for converging the elicitation process to near-optimally preferred designs. We apply the process to visual preferences for three-dimensional automobile exterior shapes. Evolutionary methods can be useful for design exploration, but learning-based methods have a stronger theoretical foundation and are more successful in eliciting subject preferences efficiently.

KW - Active learning

KW - Design preference elicitation

KW - Genetic algorithm

KW - Statistical learning

UR - http://www.scopus.com/inward/record.url?scp=84858827598&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84858827598&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781904670308

VL - 10

SP - 149

EP - 158

BT - ICED 11 - 18th International Conference on Engineering Design - Impacting Society Through Engineering Design

ER -