Byeong‐Soo Bae

About

Byeong‐Soo Bae is a rising leader in the field of mechanical metamaterials and computational inverse design, with a focus on creating materials that exhibit programmable, strain-dependent mechanical properties. His most notable contribution is the development of a Constrained Generative Inverse Design Network (CGIDN), a deep learning framework that enables the customization of metamaterials to achieve desired Poisson’s ratios with unprecedented precision. By incorporating a PCA-weighted loss function, Bae significantly improved the training efficiency and accuracy of the neural network, allowing for high-fidelity inverse design validated through both finite element analysis and physical experiments. This work, published in 2024, has already garnered 27 citations, reflecting its immediate impact on the metamaterials community. Bae’s research bridges the gap between artificial intelligence and materials engineering, offering a scalable pathway to design advanced mechanical structures for applications in robotics, aerospace, and biomedical devices. His innovative approach positions him at the forefront of data-driven materials discovery, inspiring a new generation of engineers to explore the synergy between generative models and physical design.

Research Focus

Key Achievements

Top Papers

1. 1
   Customizable metamaterial design for desired strain-dependent Poisson’s ratio using constrained generative inverse design network

   27 citations · 2024

Key Collaborators

Contact & Links