(Adam) Cheol Woo Kim

I am a postdoctoral fellow in Computer Science at Harvard University, advised by Professor Milind Tambe. My research broadly centers on AI for decision-making, with a current focus on the following areas:

• Reinforcement Learning with Human Feedback (RLHF)
  I study RLHF (primarily in the context of LLMs), with two main research directions.
  1. Multi-objectivity: I explore how to design RL policies that aggregate diverse, and often conflicting, human objectives. My current focus is on developing methods that help decision-makers understand the trade-offs and choose suitable aggregation strategies.
  2. Efficiency: I explore how to train models with minimal human supervision and reduced reliance on expensive data.
• ML-Accelerated Algorithms
  I explore how machine learning can be leveraged to efficiently solve complex optimization and control problems. I have developed ML-accelerated algorithms for mixed-integer optimization, continuous-time optimal control, and robust optimization.

Before joining Harvard, I completed my PhD at the MIT Operations Research Center in 2024 under the guidance of Professor Dimitris Bertsimas. I also spent the summer of 2023 as a research intern in the Machine Learning and Optimization Group at Microsoft Research. I completed my undergraduate studies at Seoul National University.