Jyun-Li Lin
Papers
1
Total Citations
3
H-Index
1
About
Jyun-Li Lin is a leading researcher in reinforcement learning (RL), with a particular focus on action-constrained environments where safety and resource limits are critical. His most notable contribution, "Escaping from Zero Gradient: Revisiting Action-Constrained Reinforcement Learning via Frank-Wolfe Policy Optimization" (2021), addresses a fundamental flaw in projection-based RL methods: the vanishing gradient problem. By introducing Frank-Wolfe policy optimization, Lin enables agents to learn effectively under strict constraints without sacrificing performance—a breakthrough for applications like robotic control and networked system scheduling. This work has garnered 3 citations and is recognized for its theoretical rigor and practical impact. Lin’s research bridges the gap between constrained optimization and deep RL, offering scalable solutions for real-world decision-making. His innovative approach has influenced subsequent work in safe and resource-aware AI, making him a key figure in advancing RL beyond simulated environments. For students and researchers, Lin’s work exemplifies how tackling core algorithmic challenges can unlock new frontiers in autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1