Miyabi Fujita
Papers
1
Total Citations
13
H-Index
1
About
Dr. Miyabi Fujita is a pioneering researcher at the intersection of hypercomplex neural networks and advanced control systems. Her work centers on extending neural network architectures beyond real-valued domains into higher-dimensional algebras, with a particular focus on octonion-valued systems. Her most cited paper, "Remarks on Octonion–valued Neural Networks with Application to Robot Manipulator Control" (2021, 13 citations), introduces a novel framework that leverages the unique algebraic properties of octonions to model complex, high-dimensional interactions in robotic control. This contribution is significant because it pushes the boundaries of hypercomplex neural networks—traditionally limited to complex numbers and quaternions—into the challenging, non-associative octonion domain, offering new possibilities for representing and controlling intricate robotic dynamics. Dr. Fujita’s work has been recognized for its theoretical depth and practical relevance, demonstrating how advanced algebraic structures can enhance the performance and expressiveness of neural networks in real-world engineering applications. Her research continues to inspire further exploration into hypercomplex-valued learning systems and their deployment in robotics and intelligent control.
Research Focus
Key Achievements
Top Papers
- 1