Associative spatial networks in architectural design: Artificial cognition of space using neural networks with spectral graph theory

John Harding, Christian Derix

Research output: Chapter in Book/Report/Conference proceedingChapter

  • 2 Citations

Abstract

This paper looks at a new way of incorporating unsupervised neural networks in the design of an architectural system. The approach involves looking the whole lifecycle of a building and its coupling with its environment. It is argued that techniques such as dimensionality reduction are well suited to architectural design problems whereby complex problems are commonplace. An example project is explored, that of a reconfigurable exhibition space where multiple ephemeral exhibitions are housed at any given time. A modified growing neural gas algorithm is employed in order cognize similarities of dynamic spatial arrangements whose nature are not known a priori. By utilising the machine in combination with user feedback, a coupling between the building system and the users of the space is achieved throughout the whole system life cycle.
LanguageEnglish
Title of host publicationDesign Computing and Cognition '10
Place of PublicationDordrecht
PublisherSpringer
Pages305-323
Number of pages19
ISBN (Electronic)978-94-007-0510-4
ISBN (Print)978-94-007-0509-8
DOIs
StatusPublished - 2011
Event4th International Conference on Design Computing and Cognition, DCC'10, July 12, 2010 - July 14, 2010 - Stuttgart, Germany
Duration: 1 Jan 2011 → …

Publication series

NameDesign Computing and Cognition '10
PublisherSpringer Science and Business Media

Conference

Conference4th International Conference on Design Computing and Cognition, DCC'10, July 12, 2010 - July 14, 2010
CountryGermany
CityStuttgart
Period1/01/11 → …

Fingerprint

Architectural design
Graph theory
Neural networks
Life cycle
Feedback
Gases

Cite this

Harding, J., & Derix, C. (2011). Associative spatial networks in architectural design: Artificial cognition of space using neural networks with spectral graph theory. In Design Computing and Cognition '10 (pp. 305-323). (Design Computing and Cognition '10). Dordrecht: Springer. https://doi.org/10.1007/978-94-007-0510-4_17

Associative spatial networks in architectural design : Artificial cognition of space using neural networks with spectral graph theory. / Harding, John; Derix, Christian.

Design Computing and Cognition '10. Dordrecht : Springer, 2011. p. 305-323 (Design Computing and Cognition '10).

Research output: Chapter in Book/Report/Conference proceedingChapter

Harding, J & Derix, C 2011, Associative spatial networks in architectural design: Artificial cognition of space using neural networks with spectral graph theory. in Design Computing and Cognition '10. Design Computing and Cognition '10, Springer, Dordrecht, pp. 305-323, 4th International Conference on Design Computing and Cognition, DCC'10, July 12, 2010 - July 14, 2010, Stuttgart, Germany, 1/01/11. https://doi.org/10.1007/978-94-007-0510-4_17
Harding J, Derix C. Associative spatial networks in architectural design: Artificial cognition of space using neural networks with spectral graph theory. In Design Computing and Cognition '10. Dordrecht: Springer. 2011. p. 305-323. (Design Computing and Cognition '10). https://doi.org/10.1007/978-94-007-0510-4_17
Harding, John ; Derix, Christian. / Associative spatial networks in architectural design : Artificial cognition of space using neural networks with spectral graph theory. Design Computing and Cognition '10. Dordrecht : Springer, 2011. pp. 305-323 (Design Computing and Cognition '10).
@inbook{04e20023dc8941089e59d408bd8d8c40,
title = "Associative spatial networks in architectural design: Artificial cognition of space using neural networks with spectral graph theory",
abstract = "This paper looks at a new way of incorporating unsupervised neural networks in the design of an architectural system. The approach involves looking the whole lifecycle of a building and its coupling with its environment. It is argued that techniques such as dimensionality reduction are well suited to architectural design problems whereby complex problems are commonplace. An example project is explored, that of a reconfigurable exhibition space where multiple ephemeral exhibitions are housed at any given time. A modified growing neural gas algorithm is employed in order cognize similarities of dynamic spatial arrangements whose nature are not known a priori. By utilising the machine in combination with user feedback, a coupling between the building system and the users of the space is achieved throughout the whole system life cycle.",
author = "John Harding and Christian Derix",
year = "2011",
doi = "10.1007/978-94-007-0510-4_17",
language = "English",
isbn = "978-94-007-0509-8",
series = "Design Computing and Cognition '10",
publisher = "Springer",
pages = "305--323",
booktitle = "Design Computing and Cognition '10",

}

TY - CHAP

T1 - Associative spatial networks in architectural design

T2 - Artificial cognition of space using neural networks with spectral graph theory

AU - Harding, John

AU - Derix, Christian

PY - 2011

Y1 - 2011

N2 - This paper looks at a new way of incorporating unsupervised neural networks in the design of an architectural system. The approach involves looking the whole lifecycle of a building and its coupling with its environment. It is argued that techniques such as dimensionality reduction are well suited to architectural design problems whereby complex problems are commonplace. An example project is explored, that of a reconfigurable exhibition space where multiple ephemeral exhibitions are housed at any given time. A modified growing neural gas algorithm is employed in order cognize similarities of dynamic spatial arrangements whose nature are not known a priori. By utilising the machine in combination with user feedback, a coupling between the building system and the users of the space is achieved throughout the whole system life cycle.

AB - This paper looks at a new way of incorporating unsupervised neural networks in the design of an architectural system. The approach involves looking the whole lifecycle of a building and its coupling with its environment. It is argued that techniques such as dimensionality reduction are well suited to architectural design problems whereby complex problems are commonplace. An example project is explored, that of a reconfigurable exhibition space where multiple ephemeral exhibitions are housed at any given time. A modified growing neural gas algorithm is employed in order cognize similarities of dynamic spatial arrangements whose nature are not known a priori. By utilising the machine in combination with user feedback, a coupling between the building system and the users of the space is achieved throughout the whole system life cycle.

U2 - 10.1007/978-94-007-0510-4_17

DO - 10.1007/978-94-007-0510-4_17

M3 - Chapter

SN - 978-94-007-0509-8

T3 - Design Computing and Cognition '10

SP - 305

EP - 323

BT - Design Computing and Cognition '10

PB - Springer

CY - Dordrecht

ER -