OUR LAB FOCUSES ON...

Our research focuses on developing robust machine learning (ML) frameworks tailored for limited-data scenarios in materials science and chemical engineering. Due to the high cost and limited availability of experimental data, we aim to construct prediction models that extract maximum value from small datasets. By leveraging techniques such as active learning, transfer learning, data augmentation, and ensemble modeling, we improve model's prediction accuracy and confidence even in data-sparse environments. Our approach offers the potential to significantly accelerate formulation and processing optimization while reducing the time and expense associated with conventional trial-and-error experimentation.