Doseok Jang
Papers
1
Total Citations
3
H-Index
1
About
Doseok Jang is a rising researcher at the intersection of reinforcement learning (RL) and building control systems, with a focus on developing robust, generalizable AI solutions for real-world infrastructure. His most-cited work, "Active Reinforcement Learning for Robust Building Control" (2024, 3 citations), tackles a critical challenge in RL: the brittleness of agents that overfit to training environments and fail to generalize. Jang proposes an active learning framework that enables RL agents to adaptively seek out informative experiences, improving their resilience when deployed in novel building settings. This contribution is pivotal for the practical adoption of AI in energy-efficient building management, where reliability across diverse conditions is paramount. By addressing the gap between simulated training and real-world deployment, Jang’s research promises to make smart building control both more robust and more scalable. His work stands out for its focus on bridging theoretical RL advances with tangible engineering applications, positioning him as a key voice in the future of autonomous infrastructure and sustainable urban systems.
Research Focus
Key Achievements
Top Papers
- 1Active Reinforcement Learning for Robust Building Control3 citations · 2024