Max Robotics — Global Service Robot Directory + US Certification

About

Tomoki Hamagami is a researcher at the forefront of merging computational intelligence with autonomous robotic systems. His primary research areas span complex-valued reinforcement learning, neural network evolution, and intelligent assistive robotics. Hamagami’s most notable contribution is the pioneering "Complex-Valued Reinforcement Learning" algorithm (2006, 12 citations), which introduced phase and amplitude representations from complex-valued neural networks into reinforcement learning—a novel approach that enhances context-based decision-making in partially observable environments. He has also made significant strides in developing intelligent wheelchairs (IWC) that achieve autonomous, cooperative, and collaborative behavior through state space partitioning and sensor alignment. In evolutionary computation, Hamagami proposed modified gene coding for block-based neural networks that consider failure tolerance, and developed a Hebbian learning rule for pulse neural networks that effectively restrains catastrophic forgetting. His work demonstrates a consistent focus on creating robust, adaptive learning systems that can operate reliably in real-world conditions, particularly for assistive technologies. With a career spanning foundational papers in complex-valued learning and practical robotic implementations, Hamagami’s research continues to influence the intersection of neural computation and autonomous systems.