Ganghun Lee
Papers
1
Total Citations
10
H-Index
1
About
Ganghun Lee is a pioneering researcher at the intersection of robotics, artificial intelligence, and creative automation. His primary research areas include hierarchical reinforcement learning, robotic manipulation, and stroke-based rendering for autonomous artistic agents. Lee’s most notable contribution is the development of a deep decoupled hierarchical reinforcement learning framework that enables a robotic sketching agent to simultaneously learn stroke-based rendering and precise motor control—a breakthrough that bridges high-level artistic intent with low-level robotic actuation. His landmark 2022 paper, "From Scratch to Sketch," has garnered 10 citations and represents a foundational step toward machines that can autonomously create freehand sketches, moving beyond pre-programmed trajectories. This work has significant implications for human-robot collaboration in creative fields, assistive art technologies, and adaptive manufacturing. Lee’s research demonstrates how complex, multi-scale decision-making problems can be decomposed into manageable sub-policies, offering a scalable paradigm for teaching robots sophisticated, real-world skills. His contributions are inspiring a new generation of researchers exploring the synergy between deep reinforcement learning and embodied creative intelligence.
Research Focus
Key Achievements
Top Papers
- 1