id |
ascaad2016_047 |
authors |
Algeciras-Rodríguez, José |
year |
2016 |
title |
Trained Architectonics |
source |
Parametricism Vs. Materialism: Evolution of Digital Technologies for Development [8th ASCAAD Conference Proceedings ISBN 978-0-9955691-0-2] London (United Kingdom) 7-8 November 2016, pp. 461-468 |
summary |
The research presented here tests the capacity of artificial-neural-network (ANN) based multi-agent systems to be implemented in architectural design processes. Artificial Intelligence algorithms allow for a new approach to design, taking advantage of its generic functioning to produce meaningful outcomes. Experimentation within this project is based on Self-Organizing Maps (SOMs) and takes advantage of its behavior in topology to produce architectural geometry. SOMs as full stochastic processes involve randomness, uncertainty and unpredictability as key features to deal with during the design process. Following this behavior, SOMs are used to transmit information, which, instead of being copied, is reproduced after a learning (training) process. Pre-existent architectural objects are taken as learning models as they have been considered masterpieces. In this context, by defining the SOM input set, masterpieces become measurement elements and can be used to set a distance to the new element position in a comparatistic space. The characteristics of masterpieces get embedded within the code and are transmitted to 3D objects. SOM produced objects from a population with shared characteristics where the masterpiece position is its probabilistic center point. |
series |
ASCAAD |
email |
|
full text |
file.pdf (423,439 bytes) |
references |
Content-type: text/html
Access Temporarily Restricted
Access Temporarily Restricted
Too many requests detected. Please wait 60 seconds or verify that you are a human.
If you are a human user and need immediate access, you can click the button below to continue:
If you continue to experience issues, please open a ticket at
papers.cumincad.org/helpdesk
|
last changed |
2017/05/25 13:33 |
|