Pengyi Li
Papers
1
Total Citations
10
H-Index
1
About
Pengyi Li is a rising researcher in artificial intelligence, with a primary focus on reinforcement learning (RL) and its application to complex, real-world control problems. His most notable contribution is the development of **HyAR**, a novel framework for addressing **discrete-continuous hybrid action spaces**—a common yet challenging scenario in domains like robot control and game AI. Prior to Li’s work, most RL methods were designed exclusively for either discrete or continuous actions, leaving a critical gap in handling tasks that require both, such as selecting a target (discrete) and adjusting a speed (continuous). HyAR introduces a hybrid action representation that seamlessly bridges this divide, enabling more efficient and effective policy learning. With over 10 citations since its 2021 publication, this work has already garnered attention for its practical significance. Li’s research stands out for its focus on bridging theoretical RL advances with real-world deployment, and his work on HyAR is a key step toward more versatile and intelligent autonomous systems. As he continues to explore hybrid decision-making, his contributions are poised to influence both academic research and applied AI.
Research Focus
Key Achievements
Top Papers
- 1