Yujing Hu
Papers
1
Total Citations
34
H-Index
1
About
Yujing Hu is a leading researcher in artificial intelligence, specializing in multi-agent reinforcement learning (MARL) and its applications to complex, real-world systems. His most influential work, "Value Function Transfer for Deep Multi-Agent Reinforcement Learning Based on N-Step Returns" (2019, 34 citations), addresses a critical challenge in MARL: accelerating learning in sparse-interaction environments, such as robot control and soccer games. Hu pioneered a method to transfer single-agent knowledge—specifically value functions—to multi-agent settings, enabling agents to learn more efficiently by leveraging prior experience. This approach overcomes the limitations of traditional bisimulation-based techniques, offering a scalable solution for systems where agents interact only occasionally. His contributions have advanced the practical deployment of MARL in robotics, autonomous coordination, and game AI. With a citation count reflecting growing influence, Hu’s work is foundational for researchers seeking to bridge single-agent and multi-agent learning paradigms. His innovative use of n-step returns to improve knowledge transfer has made him a key figure in the evolution of deep reinforcement learning, inspiring further exploration into efficient, transferable AI systems.
Research Focus
Key Achievements
Top Papers
- 1