We introduce a theory-driven mechanism for learning a neural network model that performs generative topology design in one shot given a problem setting, circumventing the conventional iterative process that computational design tasks usually entail. The proposed mechanism can lead to machines that quickly respond to new design requirements based on its knowledge accumulated through past experiences of design generation. Achieving such a mechanism through supervised learning would require an impractically large amount of problem–solution pairs for training, due to the known limitation of deep neural networks in knowledge generalization. To this end, we introduce an interaction between a student (the neural network) and a teacher (the optimality conditions underlying topology optimization): The student learns from existing data and is tested on unseen problems. Deviation of the student's solutions from the optimality conditions is quantified, and used for choosing new data points to learn from. We call this learning mechanism "theory-driven", as it explicitly uses domain-specific theories to guide the learning, thus distinguishing itself from purely data-driven supervised learning. We show through a compliance minimization problem that the proposed learning mechanism leads to topology generation with near-optimal structural compliance, much improved from standard supervised learning under the same computational budget.

Ruijin Cang

Hope Yao

Yi Ren

Computer-Aided Design

COMPUTER-AIDED DESIGN 是一份领先的国际期刊，为学术界和工业界提供有关计算机在设计中的应用研究和发展的重要论文。COMPUTER-AIDED DESIGN 邀请报告新研究以及新颖或特别重要应用的论文，广泛的主题，涵盖从概念创建到制造等设计过程的所有阶段。

CAD Computer Aided Design is a leading international journal that provides important papers on the application research and development of computers in design for both academia and industry. The journal invites reports of new research and papers on novel or particularly significant applications, covering a wide range of topics throughout all stages of the design process, from concept creation to manufacturing.

《CAD Computer Aided Design》是一本领先的国际期刊，专注于计算机在设计领域的应用研究与发展。该期刊欢迎提交新研究成果以及具有新颖性或特别重要性的应用论文，涵盖从概念创作到制造等设计过程的各个阶段。

CAD Computer Aided Design

One-shot generation of near-optimal topology through theory-driven machine learning

A theory-driven mechanism for learning a neural network model that performs generative topology design in one shot given a problem setting, circumventing the conventional iterative process that computational design tasks usually entail is introduced.

Category	Quartile
计算机科学	3
计算机科学, 计算机软件工程	3

One-shot generation of near-optimal topology through theory-driven machine learning

ivySCI AI Smartly Parses PDF, Answers Researchers' Questions, and Helps You Understand Papers in Seconds

Journal Info

Category	Quartile
COMPUTER SCIENCE, SOFTWARE ENGINEERING	2