Hyeok‐jin Kwon
Papers
1
Total Citations
4
H-Index
1
About
Hyeok‑jin Kwon is a rising researcher in reinforcement learning (RL) and robotics, best known for pioneering methods that simplify reward design for complex, acrobatic robotic tasks. His most cited work, “Stage‑Wise Reward Shaping for Acrobatic Robots: A Constrained Multi‑Objective Reinforcement Learning Approach” (2025), introduces an intuitive, stage‑wise RL framework that replaces monolithic reward functions with a sequence of constrained, multi‑objective sub‑goals. This breakthrough reduces the manual tuning burden and enables robots to learn agile maneuvers—such as flips and spins—more reliably. With 4 citations already, the paper is gaining traction for its practical impact on real‑world robot control. Kwon’s broader research focuses on constrained RL, multi‑objective optimization, and reward engineering, aiming to make advanced robotic learning accessible to non‑experts. His work bridges the gap between theoretical RL advances and deployable robotic systems, positioning him as a key contributor to the next generation of autonomous, acrobatic machines.
Research Focus
Key Achievements
Top Papers
- 1