Invited Talk:    Generative AI for Materials Microstructures: from Generation to Inverse Design 

Xiaoyang Zheng, 

the University of Tokyo, Japan

Abstract.

Designing novel materials is greatly dependent on understanding the design principles, physical mechanisms, and modeling methods of material microstructures, requiring experienced designers with expertise and several rounds of trial and error.  More importantly, tailoring the microstructure arrangements enables the creation of metamaterials with unprecedented properties. However, current material microstructure design considerably relies on experienced designers' inspiration through trial and error, while investigating their mechanical properties and responses entails time-consuming mechanical testing or computationally expensive simulations. Nevertheless, recent advancements in generative artificial intelligence (AI) have revolutionized the material design process, enabling property prediction and geometry generation without prior knowledge. Furthermore, deep generative models can transform conventional forward design into inverse design.

In this presentation, I will explore the use of generative AI in designing and generating material microstructures. I will begin with applications of deep learning for predicting material properties, then move on to generating 3D material microstructures using deep generative models and large language models. Finally, I will demonstrate how generative AI can be applied to inverse design of microstructures to achieve desired material properties.

Reference:

[1]   Zheng, X., Shiomi, J., & Yamada, T. (2025). Optimizing Metamaterial Inverse Design with 3D Conditional Diffusion Model and Data Augmentation. Advanced Materials Technologies, 2500293.

[2]   Zheng, X., Watanabe, I., Paik, J., Li, J., Guo, X., & Naito, M. (2024). Text‐to‐Microstructure Generation Using Generative Deep Learning. Small, 2402685.

[3]   Zheng, X., Zhang, X., Chen, T. T., & Watanabe, I. (2023). Deep learning in mechanical metamaterials: from prediction and generation to inverse design. Advanced Materials, 35(45), 2302530.

[4]   Zheng, X., Chen, T. T., Jiang, X., Naito, M., & Watanabe, I. (2023). Deep-learning-based inverse design of three-dimensional architected cellular materials with the target porosity and stiffness using voxelized Voronoi lattices. Science and Technology of Advanced Materials, 24(1), 2157682.

[5]   Zheng, X., Chen, T. T., Guo, X., Samitsu, S., & Watanabe, I. (2021). Controllable inverse design of auxetic metamaterials using deep learning. Materials & Design, 211, 110178.