Yasutaka Hatakeyama

Papers

Total Citations

H-Index

About

Yasutaka Hatakeyama is a researcher whose foundational work in neural network architectures has contributed to the evolution of visual object recognition. His key research areas include biologically inspired computer vision, specifically focusing on the neocognitron—a hierarchical, multilayered neural network model. Hatakeyama’s major contribution, "Detecting a target object using an expanded neocognitron" (1995), advanced the original neocognitron by introducing mechanisms for more robust target detection, addressing challenges in shift-invariant pattern recognition. While this early work has garnered modest citation counts (3 citations), it represents a stepping stone in the development of deep learning paradigms that later revolutionized the field. Hatakeyama’s research underscores the importance of integrating neurophysiological principles into computational models, influencing subsequent studies on hierarchical feature extraction. His achievements lie in bridging theoretical neuroscience with practical machine learning applications, offering insights that continue to inform modern convolutional neural networks. For students and researchers exploring the roots of deep learning, Hatakeyama’s work provides a valuable historical perspective on the iterative refinement of neural architectures.