# Deep Learning Architect - Classification for Architectural Design through the Eye of Artificial Intelligence (2018)
URL: https://www.associativetrails.com/writing/2022/05/02/deep-learning-architect-classification-for-architectural-design-through-the-eye-of-artificial-intelligence-2018/
Published: 2022-05-02 · Practice · Sheet 05
Summary: Notes on a 2018 paper by Yuji Yoshimura, Bill Cai, Zhoutong Wang, and Carlo Ratti

The paper Yoshimura, Cai, Wang & Ratti · 2018

arXiv (https://arxiv.org/abs/1812.01714) ResearchGate (https://www.researchgate.net/publication/332997690_Deep_Learning_Architect_Classification_for_Architectural_Design_Through_the_Eye_of_Artificial_Intelligence) PDF 5.4 MB (https://www.associativetrails.com/assets/uploads/1812.01714.pdf)

- Convnet to classify works belonging to 34 different architects. 77% accuracy.
- Using images scraped from he web, and some photos taken by researchers
- Architects then clustered together and compared to conventional architecture theory
- Difficulty in classifying modern architects - detecting space, not features
- Uses NASNet - Zoph and Schlens (2018) (https://arxiv.org/abs/1707.07012)
- Gradient-weighted Class Activation Mapping (Grad-CAM). Enables us to understand the focus of the machine eye for classification
- Dimension reduction and clustering - Linear principal component analysis. k-means clustering
- 34 architects - all past Prisker Prize winners
- 20,000 total sample images
- Goolge Tensorflow. Two GeForce GTX 1070Ti. Training complete in 8 hours.
- 30 epochs.
- Tends to confuse Koolhaas, Holl, Perrault with other architects
- Correctly distinguishes Kahn, Siza & van de Rohe
- “Top 1 accuracy indicates the probability of whether the image can correctly match with the target label. Conversely, the top 5 accuracy suggests the probability of whether the correct image can appear with the target label among five pictures ordered according to their highest probability.”
- Almost 70% of architects can be distinguished with more than 80% probability (Top 5 accuracy)
- 45% distinguished with 90% probability (Top 5 accuracy)
- Indoor scenes more distinguishable to machine eye (~5% more accurate)
- k-means clustering:
- Foster, Renzo, Rogers (High-Tech Design)
- Lloyd Wright, "Normal House" (Prarie style)
- Ando, Eisenman, Miralles (Deconstructivists)
- Mayne, Gehry

- Does not distinguish Architect's work over time