Jun Sakuma
Papers
2
Total Citations
16
H-Index
2
About
Jun Sakuma is a leading researcher in reinforcement learning and evolutionary computation, with a focus on bridging the gap between theoretical control and practical robotic applications. His major contributions center on **instance-based policy learning** and **direct policy search (DPS)** using real-coded genetic algorithms, particularly for complex, nonholonomic systems that resist mathematical modeling. Sakuma’s work demonstrates how optimization techniques can derive effective control policies where traditional methods fail, offering a robust alternative for stabilization and control in robotics. His most-cited paper (2009, 13 citations) pioneers an instance-based approach to stabilize nonholonomic systems, while his foundational 2008 study (3 citations) formalizes the optimization of such policies through genetic algorithms. Though citation counts are modest, his research is notable for its practical impact on real-world control problems, blending reinforcement learning with evolutionary optimization to solve intractable tasks. Sakuma’s achievements highlight the power of data-driven policy search in advancing autonomous systems, making his work a valuable reference for students and researchers exploring adaptive control and evolutionary robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2