Artificial Agency explores the correlations between Artificial Intelligence and algorithmic generative design, focusing on seeking emerging design approaches. The project is inspired by the explosive development of Deep Learning, which speculates on the enormous impacts of the technology on architecture.

The project is structured by a set of research methods. Initially, algorithmic mechanisms of multiple Machine Learning (ML) frameworks are analyzed and reflected regarding their potential relevance with generative design methodology. Furthermore, the research attempts to establish an emerging design procedure that embeds the most potential ML approach with existing generative systems. The posited design approaches are then applied and examined for validation and optimization in design experiments. 

As the most recent research achievements, the research develops a Reinforcement Learning based design approach that enhance generative design processes to achieve unprecedented performances and capacities. Different from existing algorithmic optimization approaches, it trains generative systems to form a specific artificial intuition toward sophisticated design intentions and unpredictable scenarios. Additionally, the research reveals a set of future directions to expand the posited approaches in robotic fabrications.

The research in Artificial Agency contributes to broadening the boundary of generative design’s existing knowledge framework, which also has significance in bridging AI and architecture discipline.

PUBLICATIONS

• Wang, D., Snooks, R. (2021). Artificial Intuitions of Generative Design: An Approach Based on Reinforcement Learning. In: Yuan, P.F., Yao, J., Yan, C., Wang, X., Leach, N. (eds) Proceedings of the 2020 DigitalFUTURES. CDRF 2020. Springer, Singapore. https://doi.org/10.1007/978-981-33-4400-6_18

• Wang, D., Snooks, R. (2021). Intuitive Behavior - The Operation of Reinforcement Learning in Generative Design Processes. In: Proceedings of the 26th Conference on Computer Aided Architectural Design Research in Asia (CAADRIA)