Machine learning for drug discovery
Bio: I am an assistant professor in the Department of Computer Science and Engineering and Graduate School of Artificial Intelligence, and a member of the Machine Learning Lab at POSTECH. Prior to this, I was a postdoctoral research associate under the supervision of Prof. Le Song and Prof. Eric Xing at MBZUAI. I received my Ph.D. under my advisor Prof. Jinwoo Shin at KAIST. My research revolves around machine learning problems with graph-structured data and combinatorial optimization problems. This includes many interdisciplinary research like applying machine learning to solve drug design, retrosynthesis, and even the maximum independent set problem from mathematics.
Abstract: Drug discovery is a very time-consuming and expensive process; fully developing a new drug takes around 10 years and $2.5B nowadays. In this talk, I will discuss how this drug discovery problem is a new and exciting field to design machine learning algorithms for structured data. In particular, I will introduce three important machine learning problems for drug discovery: molecular property prediction, molecule design, and retrosynthesis. These problems offer us the opportunity to study machine learning for graphs, 3D coordinates, combinatorial optimization, and path-finding algorithms. I will briefly go over my perspectives and some of my recent works on these problems.
-Meeting ID: 977 5406 0329