"Machine Understanding of Architectural Space" with Mikhael Johanes
Mikhael Johanes is a Ph.D. candidate at the EPFL Media and Design Laboratory, specializing in AI-driven design and machine learning applications in architecture. His research focuses on the intersection of architecture, machine learning, and computational design, contributing to various projects, publications, and exhibitions. Notably, he participated in the 2021 Seoul Biennale with the installation "Artificial Swissness" and authored the article "On GANs, NLP, and Architecture," which received the 2023 ACSA TAD Best Article Award.
Mikhael completed both his master's and bachelor's degrees in architecture at Universitas Indonesia, Jakarta, where he continued to teach and conduct research before pursuing his Ph.D. at EPFL. In collaboration with Frederick Kim Chando Kim, he leads Creative Data and Critical Design, a research initiative exploring innovative machine learning techniques for architecture and urbanism.
PRESENTATION TOPIC:
Machine Understanding of Architectural Space
This research explores how machine learning (ML) can enhance the understanding of architectural space by focusing on isovists—spatial perception models used in architecture and environmental psychology. While most AI applications lack domain-specific insights, this study bridges that gap by developing representation learning frameworks that capture the spatial patterns of built environments. The research reveals how latent spaces—compact internal data representations—can facilitate both analytical and generative tasks in architecture. This work emphasizes the importance of aligning AI with architectural inquiry to advance the theory and practice of machine understanding in architecture.
SOCIALS:
Website: https://cdcd.ch/
Instagram: / johanes_mijo
/ cdcd.ch
LIVE WITH is series of conversations and presentations with researchers exploring the use of Ai in architecture, design, and arts.
Explore more exclusive content and insights from our captivating YouTube live talks by joining us on Instagram: instagram.com/diffusion_architecture
#diffusionarchitecture #ai #architecture #creator #artist #talk