Kimin Leeمشاهده پروفایل
استادیار
- Safe and capable decision-making agents
- Physical AI
- Reinforcement Learning from Human Feedback
- +۹ مورد دیگر
Kimin Lee is an assistant professor at the Graduate School of AI at Korea Advanced Institute of Science and Technology (KAIST), where he focuses on developing safe and capable decision-making agents. His research spans multiple aspects of artificial intelligence with a strong emphasis on safety and reliability. Dr. Lee completed his educational journey at KAIST, earning a Ph.D. in Electrical Engineering with a focus on Machine/Deep Learning (2015-2020), advised by Professor Jinwoo Shin. He also holds a Master's degree in Electrical Engineering (Wireless Communication Networks, 2013-2015) and a Bachelor's degree in Electrical Engineering (2009-2013), both from KAIST. His primary research interests include: Physical AI - developing AI systems that can interact safely and effectively with the physical world Alignment - particularly reinforcement learning from human feedback (RLHF) and scalable oversight techniques Monitoring - safety evaluation frameworks and benchmarking for AI systems LLM Agents - enhancing the capabilities and safety of large language model-based agents Dr. Lee's recent publications reveal a strong trajectory toward addressing critical challenges in AI safety. His work consistently bridges theoretical advances with practical applications, particularly in the areas of reinforcement learning, computer vision, and natural language processing. A notable trend in his research is the development of methods to evaluate and enhance the safety of AI systems, especially large language models and diffusion models, while maintaining or improving their capabilities. As an active member of the academic community, Dr. Lee serves as an area chair for major conferences including NeurIPS, ICLR, and ICML, and regularly reviews for top-tier AI venues. He has also organized workshops focused on safe and trustworthy AI agents. Dr. Lee's research group at KAIST appears to focus on AI safety and decision-making, with research projects spanning from theoretical foundations to practical implementations of safe AI systems. His collaborative work with institutions like UC Berkeley and Google Research demonstrates the interdisciplinary nature of his research approach.
