This presentation examines how machine learning and AI can be applied to design processes, emphasizing the importance of data as a design material. It discusses various data types and how they influence the choice of analytical and generative models. The material explores essential concepts like data preprocessing (scaling, handling missing values, outlier removal), dimensional reduction (PCA, t-SNE), and the crucial role of data splitting (training, validation, testing) to create generic and robust models. Different machine learning problems such as regression, classification, and clustering are illustrated with examples, along with techniques like ensemble modeling and various neural network architectures (dense, convolutional, recurrent).

https://namjulee.github.io/njs-lab-public/work?id=2025-introductionToDesignComputation