Description:
Key Words:
Artificial Intelligence,Generative Adversarial Networks,Architectural Design,
Required Skills:
Experience with 3D Modeling Software such as Rhino is a Plus, but not necessary
Required Software:
Rhinoceros 7
Required Hardware:
Windows PC Required
Maximum number of participating students:
20
This incremental 5-day workshop will cover a series of hands-on machine learning techniques, including but not limited to 2D Style Transfer, 2.5D Generative Models and 3D Generative Models with 3D-VAE-GAN. The workshop will be project based, where participants will be asked to develop or customize generative architectural designs based on provided sample workflows.
This workshop will provide an introduction to Machine Intelligence(MI)—Artificial Intelligence and Machine Learning—in the field of Architecture, Design and Art. Through a series of lectures and hands-on workshops, participants will be given a historical introduction to MI-aided design, offered a view of the current state-of-the-art in terms of knowledge and tools, and a perspective on its future potential illustrated through real-world projects.
We will also examine the various types of machine intelligence that are applied as cutting-edge generative architectural design techniques since the invention of the concept “Computer-Aided Architectural Design” (Mitchell Williams 1975). By transplanting similar theoretical frameworks from computational cognitive neuroscience (David Marr 1982), we propose the universal computational model of generative architectural design as the approach towards creating architectural design machines that can possibly pass certain categories of Turing Tests (Alan Turing 1950). We will also invite guest lecturers from the Harvard Graduate School of Design to discuss both practical and theoretical applications of Machine Intelligence in Architectural Design.
More specifically, we will cover the following topics for architecture and Machine Intelligence:
1. How to use RunwayML for image generation, style transfer
2. How to connect RunwayML to Rhino-Grasshopper
3. How to generate 2.5D/3D shape using image processing and image lofting techniques
4. How to use prepare dataset, training with 3DGAN
5. How to deploy 3DGAN model with Flask and connect with Rhino-Grasshopper
6. How to use 3DGAN for shape synthesis in voxel format
7. How to use 3DGAN for real-time inference and visualization with Latent Walk
8. Gain basic understanding of reinforcement learning and computational congnitive science
9. Learn how machine intelligence is used in the AEC industry