Kunyyu Peng
Papers
1
Total Citations
2
H-Index
1
About
Kunyu Peng is a researcher focused on advancing autonomous learning in robotics, particularly through reinforcement learning and imitation learning frameworks. His most-cited work, "Humanoid action imitation learning via boosting sample DQN in virtual demonstrator environment" (2016), addresses a critical bottleneck in robot autonomy: the high cost and time required to collect training samples. Peng proposes a novel approach that combines Deep Q-Networks (DQN) with sample boosting techniques in a virtual demonstrator environment, enabling humanoid robots to learn complex actions through imitation with significantly reduced data requirements. This contribution is especially relevant to modern industrial automation, where efficient, sample-efficient learning can accelerate the deployment of autonomous systems. While his citation count remains modest, Peng's work tackles a fundamental challenge in robotics—bridging the gap between simulation and real-world learning. His research holds promise for making autonomous learning more practical and accessible, particularly in applications where large-scale data collection is infeasible.
Research Focus
Key Achievements
Top Papers
- 1