Kei Terayama

Kyoto University, Yokohama City University

Papers

Total Citations

152

H-Index

About

Kei Terayama is a leading researcher at the intersection of machine learning, materials science, and automated discovery. His primary research focuses on developing computational frameworks to accelerate the discovery and optimization of chemical compounds and materials. Terayama’s most impactful contribution is his work on black-box optimization, where he formulated the trial-and-error process of materials design as a formal optimization problem, enabling automated discovery. This seminal paper has garnered 132 citations, underscoring its influence in the field. He has also pioneered machine-learning techniques for efficient phase diagram construction, developing the PDC package that uses uncertainty sampling to reduce experimental time. Additionally, Terayama has advanced automated Rietveld refinement by creating robotic process automation (RPA) systems for the RIETAN-FP program, minimizing human intervention in complex crystallographic analyses. His work bridges the gap between computational algorithms and practical materials experimentation, making him a key figure in the push toward autonomous materials discovery. By integrating AI with high-throughput experimental workflows, Terayama is shaping a future where machine learning drives the efficient design of next-generation materials.