Yaocheng Zhang
Papers
1
Total Citations
11
H-Index
1
About
Yaocheng Zhang is a rising researcher at the intersection of robotics, reinforcement learning, and generative modeling. His work focuses on bridging the gap between advanced machine learning techniques and real-world robotic control, particularly through the lens of offline reinforcement learning. Zhang’s major contribution lies in stabilizing diffusion models for robotic control, addressing a critical challenge: while diffusion models excel at representing complex distributions and imitating diverse behavioral trajectories, they often struggle with feasibility and safety in deployment. In his highly cited 2024 paper, "Stabilizing Diffusion Model for Robotic Control With Dynamic Programming and Transition Feasibility," Zhang introduces a novel framework that integrates dynamic programming principles to ensure that generated actions are not only realistic but also physically and transitionally feasible. This work has quickly garnered 11 citations, signaling its impact on the field. By tackling the instability of diffusion-based policies, Zhang is paving the way for more reliable, data-driven robotic systems. His research is particularly valuable for students and engineers seeking to harness generative models for control tasks without sacrificing robustness or safety.
Research Focus
Key Achievements
Top Papers
- 1