Notes on the Foundations of Reinforcement Learning and LLM Post-Training (Part 1)
Part 1 covers RL fundamentals, policy gradients, REINFORCE, actor-critic methods, TRPO, and PPO.
Hi! I’m Chi, and I am interested in developing multimodal AI systems that can meaningfully improve medical understanding and support clinical decision-making.
My research lies at the intersection of machine learning and medical imaging, with a particular focus on multimodal foundation models, medical reasoning, and trustworthy AI for healthcare. I am especially interested in building clinically grounded and reliable models that can learn from complex medical data while remaining interpretable and useful in real-world settings.
During my research journey, I have been fortunate to work under the guidance of Prof. Yueming Jin at the National University of Singapore, Prof. Lee Hwee Kuan at the Bioinformatics Institute (A*STAR), Prof. Hieu Pham at VinUniversity, and Prof. Ravishankar K. Iyer at the University of Illinois Urbana–Champaign. Feel free to reach out if you would like to chat about medical AI, multimodal learning, or research in general!
Ph.D. Student in Computer Engineering
2025-08-01
National University of Singapore
Bachelor of Computer Science
2021-09-01
2025-06-01
VinUniversity
CVPD Lab, A*STAR Bioinformatics Institute
VinUni-Illinois Smart Health Center
Coordinated Science Lab, University of Illinois at Urbana-Champaign
National University of Singapore
VinUniversity
Part 1 covers RL fundamentals, policy gradients, REINFORCE, actor-critic methods, TRPO, and PPO.
Part 2 covers RL in the LLM/VLM setting, RLHF, the PPO training pipeline, and GRPO.