NTU-CEE Seminar Series: Associate Professor Stephen Wu

Organized By

CEE Seminar Committee

Host By

Assistant Professor Shi Chao

Topic

Role of Geotechnics Researchers in the Era of Rapid AI Advancement

About the Seminar

The rapid evolution of artificial intelligence is reshaping scientific inquiry across disciplines, and geotechnical engineering is no exception. As foundation models and large-scale pretraining paradigms emerge, geotechnics researchers face new opportunities—and responsibilities—in navigating this transformation.

This talk explores the evolving role of geotechnical researchers in the age of AI, with a focus on leveraging foundation models, transfer learning, and multiagent intelligent systems. I will examine how transfer learning from pretrained deep neural networks can mitigate the challenge of sparse, heterogeneous data commonly found in geotechnics, while highlighting the risks of domain mismatch and negative transfer. Scaling laws are discussed as a tool for estimating model efficiency and transfer viability.

Furthermore, we explore the role of large language models (LLMs) not merely as text generators, but as autonomous agents embedded in multi-agent systems capable of reasoning, planning, and collaborating in complex geotechnical workflows.

Finally, we advocate for interdisciplinary research models, community-driven benchmark datasets, and robust evaluation frameworks to ensure that geotechnical AI advances remain grounded in domain knowledge. This vision challenges researchers to move beyond adopting tools passively and instead take leadership in co-creating next-generation AI systems tailored to the demands and complexities of geotechnical engineering.

About the Speaker

Dr. Stephen Wu is an Associate Professor at the Institute of Statistical Mathematics (ISM) in Tokyo, Japan. His research focuses on Bayesian inference, machine learning, and uncertainty quantification, with applications spanning polymer science, chemoinformatic, earthquake engineering, and geotechnical engineering. He completed his M.S. and Ph.D. in Mechanical and Civil Engineering at the California Institute of Technology, where his doctoral research centered on developing probability-based earthquake early warning system and its applications under the guidance of Prof. James L. Beck and Prof. Thomas Heaton.

Following his Ph.D., Dr. Wu held postdoctoral positions at ETH Zurich's Computational Science & Engineering Laboratory, focusing on hierarchical Bayesian models for molecular dynamics and pharmacokinetics, and at ISM, working on applying latest AI technologies to diverse science and engineering applications. In recognition of his contributions to statistical machine learning and data-driven science, Dr. Wu received the Young Scientists' Award from Japan's Ministry of Education, Culture, Sports, Science and Technology in 2023.

Registration

Registration is not required, and ALL are welcome to attend.