Junyao Shi

Papers

Total Citations

H-Index

About

Junyao Shi is a robotics researcher whose work sits at the intersection of machine learning, control systems, and human-robot interaction. His primary research areas include deep reinforcement learning for locomotion and active learning from human feedback. Shi’s most notable contribution is in the design of gaits for underactuated snake robots, where he demonstrated that deep reinforcement learning can generate effective locomotion strategies that are often unintuitive and difficult to derive through traditional trial-and-error or simplified models. This work, published in 2020, has garnered 10 citations and highlights the potential of continuous state-action reinforcement learning for complex robotic systems. In parallel, Shi has advanced the field of Learning from Human Feedback by integrating active learning techniques to maximize the efficiency of non-expert human input during robot training. His 2020 paper on this topic, with 8 citations, addresses a critical bottleneck in human-in-the-loop systems: reducing the need for constant human attention while maintaining learning performance. By optimizing when and how feedback is solicited, Shi’s work paves the way for more practical and scalable robot learning from everyday users.