Koichi Sugiyama
Papers
2
Total Citations
4
H-Index
1
About
Koichi Sugiyama is a researcher in robotics and artificial intelligence, with a primary focus on reinforcement learning for autonomous systems. His work centers on bridging the gap between simulated and real-world environments, enabling robots to acquire complex control rules through trial and error without the prohibitive cost of physical experimentation. Sugiyama’s major contribution lies in developing a virtual space learning framework that allows robots to practice and refine tasks—such as manipulation or navigation—in simulation before transferring those skills to real space. This approach addresses a critical bottleneck in robotics: the high number of trials needed for reinforcement learning in physical settings. His most-cited paper, “Robot Task Learning based on Reinforcement Learning in Virtual Space” (2007), has garnered 3 citations, while his follow-up work in 2010 extends this methodology. Though his citation counts are modest, Sugiyama’s research is foundational for efficient robot skill acquisition, offering a scalable pathway to deploy intelligent machines in dynamic, real-world tasks. His work is particularly relevant for students and researchers exploring sim-to-real transfer in embodied AI.
Research Focus
Key Achievements
Top Papers
- 1Robot Task Learning based on Reinforcement Learning in Virtual Space3 citations · 2007
- 2