Encoded Memory V1.0
SCROLL DOWN
SCROLL DOWN
 
 
ENCODED MEMORY V1.0
 
 
DESCRIPTION
Machine learning techniques have developed tremendously over the last two decades. We’re currently witnessing the renaissance of the supervised learning, which is directly influencing our lives – mostly without our awareness – via the search engines, recommendations systems, social networks and more. As with any well developed technology, we are asking the same question again – how can it influence our culture and change our perspective on the surrounding world? How can we adapt those tools to be used in architecture? What could be the benefits for the designers and how can machine learning change the way we work and create? The premise of a flexible and ever-learning, ever-adapting work environment seems like a viable vision of the future for the creative industry. To achieve that, we need to know exactly how and why machine learning works. Finding a creative application for this set of tools in not an easy task, and requires us to develop a new understanding of the digital modelling concept. Our parametric models have to account both for human and machine readability, to be able to benefit from the two worlds. To make it happen we need to work on the translation between the organic and the digital memory, the common encoding.
 
 
DETAILS
Encoded Memory V1.0 is a Grasshopper workshop which aims to explore the potential of machine learning neural networks in the development of an urban fabric of a particular sector in Dubai. We will demonstrate how the parametric model is presented to the network and how does it influence its effectiveness and decisions making process in directing the design workflow. We will be testing multiple data input and will explore the network response based on the self-learning generative nature of the network. Based on the network interpretation to the input data, participants will be able to visualize the network growth outcome, and understand its data interpretation and decision making process and be able to intervene to adjust these data, or the outcome, to influence the network future direction. Behavior of the system will therefore develop by assessing their propagation on a basic organization level, then developing it to work on a variety of urban levels. 
During this workshop we will investigate the applications of neural networks in multiple roles – from the designer’s adviser to the self-learning, generative method. We will try to employ those methods in our decision-making process, learning how they can assist us in the exploration of the parametric space. In order to utilize the machine learning tools, we need to know exactly how they work. To do that, we will closely monitor the neural network training and decision making process, and learn how it responds to the data. We will be using Rhino/grasshopper with selected plug-ins such as Owl (yet to be released to public) will be used for the generation of the neural network, based on the input data from each participant or a group. The selected data and their connections reflect the design intent as set by each group. The network will process these data and will propose different layouts as it propagates the field. Participants will be able to visualize the propagation, and to guide the outcome to reflect their design intent.
 
 
PRE-REQUISITE SKILLS
The Workshop materials are structured to cater for participants with BASIC grasshopper and Rhino skills ONLY. NO scripting skills required.
Outcome: Participants will be working in groups during the last two days of the workshop, each developing an urban scheme for a selected site in Dubai. Each group will develop their own concept based on the techniques taught in the previous three days, directly supervised by workshops’ tutors.
 
 
WHAT YOU WILL LEARN
  • Machine learning concepts and notions (supervised vs unsupervised learning, neural networks, clustering etc.)
  • Application of machine learning in expanding the design space, and generating different possibilities based on the design intent.
  • Engaging with the machine learning process, and guiding the possible outcomes.
  • Data preparation for machine learning – ways of model parametrization

  • Owl Machine learning plug in for Grasshopper 3D
  • Scripting with Owl
  • Accord framework basics
 
REQUIRED SOFTWARE
Each Participant should provide their own computer with the following software installed. If you do not have a software, trial versions can be found on each software company’s website. We will send participants links to the required software, trials, and plug-ins for download. Please have the software installed prior to workshop start. If you have issues installing software, we may be able to help you during the workshop.
  • Rhinoceros
  • Grasshopper   (Plug-ins: Owl – yet to be published)
  • Accord.NET Framework
 
LOGISTICS
Encoded Memory V1.0 will expand from 04 – 08 April, 2017.
Daily meetings will take place from 10:00 to 18:00 at Dubai Knowledge Village
There will be a final presentation of your work containing renderings and diagrams on 08 April, 2017.
All projects will be published on DesignMorphine’s web page.
All participants who complete the workshop will receive a certificate of completion.
For attending the workshop there is no previous software experience required.
Participants need to bring their own laptops or workstations.
Please note that places are very limited on a first pay first serve basis!
 
PROJECT SITE
Encoded Memory V1.0

 

TUTORS

Encoded Memory V1.0
Mateusz Zwierzycki
Poland
 
Member @ DesignMorphine
Research Assistant @ CITA
MArch ASP
Poznań University of Arts, Poland
 
Graduated from the Department of Architecture and Design of the Poznań University of Arts in 2012. Focused on parametric and generative design, which he considers to be a natural way of design thinking. Computational design popularizer, tutor for many international workshops, co-author of the first Polish parametric design oriented website (projektowanieparametryczne.pl ). He developed a variety of plugins: Anemone, Starling, Squid and other scripts and freeware software for the Grasshopper community. In his spare time, he is a geometry enthusiast (awesomeshapes.tumblr.com ).

 

Encoded Memory V1.0
Zayad Motlib
United Arab Emirates
 
Founder @ D-NAT
Head of Design @ Al-Nhayan Design Office
 

Zayad is an interdisciplinary architect, a designer, and a researcher on information- based digital design and material systems. He worked in internationally renowned practices in New Zealand, Australia, and the Middle East such as Woods Bagot and LAVA on a variety of award-winning projects. Parallel to his professional practice, he has taught and served as a guest critic at several universities in Sydney, New Zealand, and across the UAE. In 2013, he has founded d-NAT (Dubai Network for Art, Architecture, and Technology); a community based network set as a hub for the creative minds to connect and to bridge the boundaries of Architecture, Design, and Technology.

Dubai Knowledge Village