Zhibo Zhou
Papers
2
Total Citations
40
H-Index
2
About
Zhibo Zhou is a researcher specializing in deep reinforcement learning (DRL) and robotic locomotion control, with a particular focus on the complex challenges of bipedal robot movement. His most recognized contribution centers on reward-adaptive reinforcement learning, a novel framework that addresses one of the fundamental difficulties in training bipedal robots: balancing multiple competing optimization criteria simultaneously. Zhou's key innovation, dynamic policy gradient optimization, introduces a mechanism for adaptively weighting rewards from different criteria during training, enabling more robust and efficient learning for non-statically stable bipedal systems. His flagship work, "Reward-Adaptive Reinforcement Learning: Dynamic Policy Gradient Optimization for Bipedal Locomotion," has accumulated 38 citations since its 2022 publication, reflecting meaningful traction within the robotics and machine learning communities. The research bridges the gap between simulation-based DRL and physical robot deployment — a notoriously difficult challenge given real-world dynamics. By demonstrating effectiveness across both simulated and physical platforms, Zhou's work contributes practically applicable solutions to legged robotics. His research is particularly relevant for students and engineers working at the intersection of reinforcement learning, control theory, and humanoid or assistive robot development.
Research Focus
Key Achievements
Top Papers
- 1
- 2