Mengbing Li

Xijing Hospital, University of Michigan–Ann Arbor

Papers

Total Citations

H-Index

About

Mengbing Li is a researcher whose work spans the intersection of advanced surgical techniques and cutting-edge machine learning methodologies. His key research areas include robotic-assisted surgery, particularly in living donor organ transplantation, and the development of robust reinforcement learning (RL) algorithms for nonstationary environments. Li made a significant contribution to transplant surgery with his pioneering work on robotic-assisted live donor ileal segmentectomy for intestinal transplantation, a minimally invasive approach designed to reduce donor morbidity. This work, published in 2017, has garnered 8 citations, highlighting its importance in advancing surgical practice. On the methodological frontier, Li addresses a critical limitation in artificial intelligence by developing tests for stationarity and change point detection in reinforcement learning. His 2022 and 2025 papers, each with 2 citations, tackle the restrictive assumption that state transitions and reward functions are constant over time, proposing offline RL methods that can adapt to dynamic environments. This dual expertise—bridging tangible surgical innovation with theoretical advances in machine learning—positions Li as a versatile researcher whose work has practical implications for both patient care and AI robustness. His contributions are particularly notable for their potential to improve safety in live donor procedures and to enhance the reliability of RL systems in real-world, nonstationary settings.