Papers
1
Total Citations
4
H-Index
1
About
Jinmin He is a rising researcher in artificial intelligence, with a primary focus on multi-task deep reinforcement learning and neural architecture design. His most notable contribution is the development of Dynamic Depth Routing, a novel framework introduced in his 2024 paper “Not All Tasks Are Equally Difficult.” This work addresses a critical challenge in multi-task reinforcement learning: how to efficiently share parameters across tasks of varying difficulty. By segmenting a neural network into distinct modules and training a routing network to dynamically recombine them, He’s approach significantly improves data efficiency and task performance. The paper has already garnered 4 citations, signaling early impact in the field. He’s research is particularly valuable for real-world applications where a single agent must master diverse skills, such as robotics and game AI. His work stands out for its elegant solution to balancing task-specific demands with shared learning, making him a promising voice in the next generation of reinforcement learning researchers.
Research Focus
Key Achievements
Top Papers
- 1