Sosuke Kobayashi
Papers
3
Total Citations
236
H-Index
3
About
Sosuke Kobayashi is a leading researcher at the intersection of robotics, natural language processing, and human-in-the-loop reinforcement learning. His work focuses on enabling robots to understand and act upon unconstrained spoken language, a critical step toward seamless human-robot collaboration. Kobayashi’s most influential contribution is his 2018 paper “Interactively Picking Real-World Objects with Unconstrained Spoken Language Instructions,” which has garnered 175 citations. This work tackles the formidable challenge of parsing complex, ambiguous spoken commands to guide robotic manipulation in real-world settings. He further advanced the field with “DQN-TAMER: Human-in-the-Loop Reinforcement Learning with Intractable Feedback” (48 citations), which integrates human feedback to overcome exploration hurdles in RL, drastically reducing the number of trials needed for robots to learn effective policies. By bridging the gap between natural language comprehension and adaptive robotic control, Kobayashi’s research has laid essential groundwork for more intuitive, interactive AI systems. His achievements highlight a commitment to making robots not just functional, but truly communicative partners.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3