Kyung Min (Brian) Ko

Research Interests: World Models, Machine Learning Research Agents, Reinforcement Learning

Welcome! I’m an M.S. student at the University of Washington. I’m fortunate to be advised by Prof. Sherry Yang. My research focuses on world models and machine learning research agents, at the intersection of large language models and reinforcement learning. I’m interested in how agents can predict the outcomes of their actions to guide experimentation and learn more efficiently. Our recent work, MLE-World, develops a learned world model that predicts execution feedback from code and task context to guide machine learning experiments.

I graduated from Purdue University with distinction. I have been fortunate to collaborate with Prof. Huan Zhang (UIUC) on the intersection of LLMs and RL, and with Prof. David I. Inouye (Purdue) on statistical machine learning.

I was grateful to be selected twice for NSF-funded summer research programs, advised by Prof. Siva Theja Maguluri (Georgia Tech) and Prof. Lin Tan (Purdue).

I’m open to collaboration—please email me if you’d like to discuss ideas.