Simple arithmetic operation in latent space can generate a novel three-dimensional graph metamaterials

Abstract

Abstract Recent advancements in artificial intelligence (AI)-based design strategies for metamaterials have revolutionized the creation of customizable architectures spanning nano- to macro-scale dimensions. However, their increasing complexity poses challenges in generating diverse metamaterials, hindering widespread adoption. Here, we introduce an innovative design strategy for three-dimensional graph metamaterials through simple arithmetic operations within the latent space. By leveraging carefully designed hidden representations of disentangled latent space and diffusion processes, our method unravels design space complexity, generating diverse graph metamaterials with comprehensive understanding. This versatile methodology facilitates the creation of graph metamaterials ranging from repetitive lattices to functionally graded materials. We believe that this methodology represents a foundational step in advancing our comprehension of the intricate latent design space, offering the potential to establish a unified model for various traditional generative models in the realm of graph metamaterials.