Chikara Utsumi
Papers
1
Total Citations
4
H-Index
1
About
Chikara Utsumi is a robotics researcher whose work lies at the intersection of deep learning and robot manipulation, with a particular focus on how human knowledge can be explicitly integrated into machine learning frameworks. His research centers on enhancing robot learning from demonstrations by leveraging symbolic human annotations, such as action labels derived from task segmentation. In his most-cited work, "Use of Action Label in Deep Predictive Learning for Robot Manipulation" (2022, 4 citations), Utsumi demonstrates that primitive human-annotated subtasks can serve as powerful guiding signals for deep predictive models, enabling robots to better understand and replicate complex manipulation sequences. This approach bridges the gap between low-level sensorimotor learning and high-level symbolic reasoning, offering a pathway toward more interpretable and data-efficient robot learning. Utsumi’s contributions are particularly relevant for advancing human-robot collaboration, where explicit human knowledge can accelerate skill acquisition. Though early in his career, his work signals a promising direction for integrating cognitive science insights into practical robotic systems.
Research Focus
Key Achievements
Top Papers
- 1Use of Action Label in Deep Predictive Learning for Robot Manipulation4 citations · 2022