Jaewon Kim
Undergraduate Student, Korea University
Learning to generate and act
in the physical world.
I am an undergraduate student in Semiconductor Engineering at Korea University, working with Prof. Junhee Seok in the Complex Data Analytics Lab (CDAL). In Spring 2027, I will join the University of Texas at Austin as an exchange student in mathematics.
My research lies at the intersection of deep reinforcement learning and generative models, with the goal of building expressive and scalable algorithms for the real world. I am particularly interested in physics-informed machine learning, which uses PDEs and stochastic processes to guide learning, and its applications to robotics and AI for science.
Research directions
What I work onExpressivity
Policies and generative models that capture multi-modal behavior instead of collapsing to a single mode.
02Scalability
Learning that stays reliable as conditions and state spaces grow high-dimensional.
03Physical consistency
Building PDEs and stochastic processes into learning so that models stay consistent with physics.