Dongge Han

Papers

Total Citations

H-Index

About

Dongge Han is a rising researcher at the forefront of reinforcement learning (RL) and robotics, with a focus on enabling intelligent agents to operate autonomously in complex, real-world environments. Her work bridges the gap between multiagent systems and human-robot interaction, addressing fundamental challenges in credit assignment and personalization. In her 2021 paper on "Multiagent Model-based Credit Assignment for Continuous Control," Han introduced novel methods for decentralized RL, allowing robotic components to learn effectively without constant communication—a critical step for scalable, real-world deployment. This work has garnered 5 citations, establishing her early impact in the field. More recently, Han's 2024 paper "LLM-Personalize" tackles the critical issue of aligning large language model (LLM) planners with individual human preferences for housekeeping robots. By leveraging reinforced self-training, she enables robots to adapt their task planning to user-specific needs, moving beyond generic, one-size-fits-all solutions. This cutting-edge research, already with 4 citations, showcases her ability to integrate LLMs with RL for practical, personalized robotics. Han’s contributions are paving the way for more adaptive, user-friendly autonomous systems, making her a notable voice in the next generation of AI-driven robotics research.