Wenjun Zeng
Papers
2
Total Citations
7
H-Index
2
About
Wenjun Zeng is a rising researcher at the forefront of reinforcement learning and robotics, whose work bridges algorithmic innovation with embodied intelligence. His primary research areas span deep reinforcement learning, exploration strategies, and the development of whole-body control systems for humanoid robots. In his most cited work, "Rewarding Episodic Visitation Discrepancy for Exploration in Reinforcement Learning" (2022, 5 citations), Zeng tackles the critical challenge of exploration in complex environments with sparse rewards. He introduces a novel intrinsic reward mechanism that leverages episodic visitation discrepancies, enabling agents to discover effective policies more efficiently—a contribution that addresses a fundamental bottleneck in deep RL. More recently, his 2025 survey, "A Survey of Behavior Foundation Model: Next-Generation Whole-Body Control System of Humanoid Robots" (2 citations), positions him at the cutting edge of robotics, offering a comprehensive framework for integrating behavior foundation models into humanoid control. This work synthesizes advances in motor control, human-robot interaction, and physical intelligence, highlighting Zeng's vision for next-generation autonomous systems. Though early in his career, his focused contributions to exploration and whole-body control signal a promising trajectory in AI and robotics research.
Research Focus
Key Achievements
Top Papers
- 1
- 2