Add To Collection
Add to Collection
The Two-Act Workflow: ML Scenarios + Rule-Based Assembly
The Two-Act Workflow: ML Scenarios + Rule-Based Assembly
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
View Original
Add To Collection
Add to Collection
The Two-Act Workflow: ML Scenarios + Rule-Based Assembly
The Two-Act Workflow: ML Scenarios is an academic research project developed at UCLA exploring how machine learning can inform architectural master planning.
Act 1 uses ML image-to-image translation to generate speculative top-view maps of Phnom Penh under different conditions—flooded delta, Venice-like canal city, and Chicago-inspired dense grid.
Act 2 transforms these 2D scenarios into a rule-based 3D assembly. Through a parametric rule table and Grasshopper scripts, prototypes are built with explicit logic, allowing reproducibility and variation through seed control.
The workflow demonstrates how speculative ML outputs can be converted into architectural intelligence: transparent, auditable, and adaptable for urban-scale design.
Rather than replacing the role of architects, ML becomes a collaborator, expanding the way we visualize and construct future cities.
Act 1 uses ML image-to-image translation to generate speculative top-view maps of Phnom Penh under different conditions—flooded delta, Venice-like canal city, and Chicago-inspired dense grid.
Act 2 transforms these 2D scenarios into a rule-based 3D assembly. Through a parametric rule table and Grasshopper scripts, prototypes are built with explicit logic, allowing reproducibility and variation through seed control.
The workflow demonstrates how speculative ML outputs can be converted into architectural intelligence: transparent, auditable, and adaptable for urban-scale design.
Rather than replacing the role of architects, ML becomes a collaborator, expanding the way we visualize and construct future cities.
Product Spec Sheet
Were your products used?
Join as a manufacturer to add your products.