Kathy Jang
Papers
1
Total Citations
13
H-Index
1
About
Kathy Jang is a rising researcher at the intersection of reinforcement learning (RL) and control theory, with a primary focus on developing robust and reliable decision-making algorithms for complex dynamical systems. Her most cited work, "Robust Reinforcement Learning using Adversarial Populations" (2020, 13 citations), makes a pivotal contribution by addressing a critical vulnerability in standard RL: its tendency to fail catastrophically under even minor perturbations to system dynamics. Jang introduced a novel formulation that trains policies against an adversarial population of dynamics models, effectively hardening the controller against worst-case scenarios. This approach bridges the gap between theoretical robustness guarantees and practical deployment, offering a pathway to safer RL applications in robotics and autonomous systems. Though early in her career, her work is already recognized for tackling the fundamental challenge of bridging simulation-trained policies to real-world uncertainty, positioning her as a promising voice in the growing field of robust and safe reinforcement learning.
Research Focus
Key Achievements
Top Papers
- 1Robust Reinforcement Learning using Adversarial Populations13 citations · 2020