Three decades of machine learning with neural networks in computer-aided architectural design (1990–2021)

Abstract Over the past years, computational methods based on deep learning—that is, machine learning with multilayered neural networks—have become state-of-the-art in main research areas computer-aided architectural design (CAAD). To understand current trends of CAAD learning, to situate them a broader historical context, and identify future challenges, this article presents systematic review publications that apply networks problems. Research papers employing were collected, particular, from CumInCad major open-access repository community categorized into different types Upon analyzing distribution these categories, namely, composition subjects, data types, network models, suggests discusses several technical trends. Moreover, it identifies analyzed typically provide limited access important components used as part their methods. The points out importance sharing training experiments data, describing dataset, dataset parameters, samples, model, results support reproducibility. It proposes guideline aims at increasing quality availability learning.