Akihito Sudo

Tokyo Institute of Technology, Shizuoka University

Papers

Total Citations

H-Index

About

Akihito Sudo is a researcher whose work lies at the intersection of neural computation and social robotics, with a particular focus on how machines learn and interact in noisy, real-world environments. His most significant contributions center on developing associative memory systems capable of robust online incremental learning. In his landmark 2009 paper, Sudo introduced a novel associative memory model using a Self-Organizing Incremental Neural Network, designed to handle the sequential, noisy data streams typical of real environments—a foundational contribution that has garnered 49 citations. This work was preceded by a 2007 paper laying the groundwork for such noise-robust, online learning mechanisms. More recently, Sudo has expanded into the modeling of initial social encounters, proposing an agent-based model in 2021 that generates approaching and avoiding behaviors based on internal states and spatial relationships, rather than pre-scripted scenarios. This shift from neural memory to social interaction dynamics highlights Sudo's versatility, exploring how computational models can simulate the subtle, non-verbal cues that initiate human communication.