Tu-Hoa Pham
Papers
3
Total Citations
135
H-Index
3
About
Tu-Hoa Pham’s research lies at the intersection of robotics, deep reinforcement learning, and constrained optimization, with a focus on enabling real-world physical interactions. His most impactful contribution is the development of OptLayer, a practical constrained optimization framework for deep reinforcement learning that allows robots to safely learn and execute complex tasks in unstructured environments—overcoming the traditional limitation of trial-and-error learning in the real world. This work, published in 2018, has garnered 118 citations, reflecting its significance in bridging the gap between simulation and physical deployment. Pham has also advanced robot learning for assembly tasks, contributing an experimental force-torque dataset for multi-shape insertion, a resource that supports the modeling of complex physical interactions such as cable stretch and fluid dynamics. His research is notable for its emphasis on practical, deployable solutions that respect real-world safety and physical constraints, making his work highly relevant for students and researchers interested in bringing reinforcement learning out of the lab and into real robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3