Seiji Ishihara
Papers
1
Total Citations
2
H-Index
1
About
Seiji Ishihara is a researcher in robotics and machine learning, with a focus on decision-making systems for autonomous agents. His work bridges reinforcement learning and robotic control, particularly in dynamic, real-world environments. His most-cited paper, "Learning of Decision Making at Free Kicks Using Policy Gradient Methods," presented at the RoboCup symposium in 2005, explores how policy gradient algorithms can enable robots to learn optimal strategies in competitive soccer scenarios. This contribution, while modest in citation count (2), reflects an early and targeted application of reinforcement learning to multi-agent coordination and tactical decision-making. Ishihara’s research addresses the challenge of enabling robots to adapt and improve their actions through experience, a cornerstone of modern autonomous systems. His work is notable for integrating RoboCup platforms—a standard testbed for AI and robotics—with policy gradient methods, offering insights into how robots can learn from sparse rewards in adversarial settings. For students and researchers, Ishihara’s studies provide a practical example of applying reinforcement learning to real-time, physical tasks, highlighting the intersection of machine learning, robotics, and game theory.
Research Focus
Key Achievements
Top Papers
- 1