Mingqi Yuan
Papers
2
Total Citations
7
H-Index
2
About
Mingqi Yuan is a researcher at the forefront of reinforcement learning and embodied AI, with a focus on enabling intelligent agents to explore and act in complex, high-dimensional environments. His work addresses the fundamental challenge of exploration in deep reinforcement learning, where sparse rewards often hinder learning. In his highly cited paper, “Rewarding Episodic Visitation Discrepancy for Exploration in Reinforcement Learning” (2022, 5 citations), Yuan introduced a novel intrinsic reward mechanism that incentivizes agents to visit novel states, significantly improving sample efficiency and performance in challenging tasks. This contribution has been recognized as a key advancement in the field. More recently, Yuan has turned his attention to the next frontier: humanoid robotics. His 2025 survey, “A Survey of Behavior Foundation Model: Next-Generation Whole-Body Control System of Humanoid Robots” (2 citations), provides a comprehensive roadmap for building general-purpose physical intelligence, synthesizing cutting-edge approaches to whole-body control. Through his work, Yuan is bridging the gap between algorithmic exploration and real-world robotic dexterity, making him a rising voice in both reinforcement learning and robotics research.
Research Focus
Key Achievements
Top Papers
- 1
- 2