Hello! My name is Hyun Kim, a fourth-year PhD candidate in transportation at New Jersey Institute of Technology.
My research interest lies in the application of Machine Learning and Reinforcement Learning.
My dissertation topic is on the usage of Deep Reinforcement Learning to dynamically route multiple shuttles based on stochastic demand.
I also have experience in data analysis, transportation modeling, and multi-agent systems.
Here you can find my research articles, projects, and publications.
The surrogate clipped objective function used in PPO is given by:
$$L^{CLIP}(\theta) = \hat{\mathbb{E}}_t \left[ \min\left(r_t(\theta)\hat{A}_t,\ \text{clip}(r_t(\theta), 1 - \epsilon, 1 + \epsilon)\hat{A}_t\right) \right]$$
A collection of my favorite landscape and travel photos.