Seunghwa Ryu

Abstract: Machine learning (ML) has transformed the mechanics of materials and manufacturing, accelerating the discovery of novel materials and enabling the development of innovative products by predicting complex processes that traditional physics-based models often struggle to capture. However, with numerous ML models available, the challenge lies in selecting the right one for specific design needs. In this seminar, I will present a schematic overview of strategies for selecting data-driven design methods to address four distinct challenges: (1) optimization through interpolation with ample training data, (2) optimization via extrapolation across vast design spaces beyond the initial dataset, (3) scenarios with limited data, and (4) multi-fidelity approaches that integrate small, precise datasets with larger, approximate ones. Drawing on examples from both academia and Korean manufacturing companies, I will explore effective data-driven optimization strategies applied to areas such as plate forming, composite structure design, Kirigami patterning, and 3D printing process optimization. As an alumnus of the Mechanics and Computation Division, I will also share my journey from computational nanomechanics to a focus on data-driven materials design and multiscale modeling, reflecting on how this transition aligns with evolving demands and opportunities in the Korean industrial landscape.

Bio: Seunghwa Ryu is a Professor of Mechanical Engineering at KAIST, where he has been a faculty member since 2013, and is currently a Visiting Professor at UC Berkeley. He earned his Ph.D. in Mechanical Engineering from Stanford University under the supervision of Prof. Wei Cai in the Mechanics & Computation Division, after completing his B.S. at KAIST. His research focuses on machine learning-based design of materials and structures, as well as computational multiscale mechanics, with a particular emphasis on applications in the manufacturing industry. Seunghwa has been honored with several awards, including the APACM Award for Young Investigators in Computational Mechanics (2019) and election to the Young Korean Academy of Science and Technology (2021). He currently serves as an associate editor of Frontiers in Materials and as an editorial advisor or board member for journals such as Scientific Reports and Advanced Intelligent Discovery. To date, he has published over 150 SCI journal papers.