Hisashi Kashima

About

Hisashi Kashima is a leading researcher in machine learning, with a primary focus on reinforcement learning and robust statistical methods. His most impactful work centers on developing algorithms that improve the reliability and efficiency of value function approximation, a core challenge in applying reinforcement learning to real-world systems like robotics. Kashima’s major contribution is the introduction of "Least Absolute Policy Iteration" (LAPI), a novel approach that replaces the traditional squared loss with an absolute loss function. This innovation significantly enhances robustness against outliers in reward signals, a common problem in noisy, real-world environments. His foundational papers on this topic, published in 2009 and 2010, have collectively garnered over 11 citations, establishing a key reference point for researchers tackling instability in policy evaluation. By addressing the sensitivity of least-squares methods, Kashima’s work provides a more reliable framework for computational learning, directly benefiting fields where data integrity is critical. His contributions underscore a commitment to bridging theoretical rigor with practical, resilient AI systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Least absolute policy iteration for robust value function approximation

   6 citations · 2009
2. 2
   Least Absolute Policy Iteration-A Robust Approach to Value Function Approximation

   5 citations · 2010

Key Collaborators

Contact & Links