Joe Nishihara

Papers

Total Citations

H-Index

About

Dr. Joe Nishihara is a leading researcher in developmental robotics and human-robot interaction, with a core focus on enabling machines to learn language and object concepts in real-time, much like humans do. His major contribution lies in pioneering unsupervised, online learning algorithms that allow robots to simultaneously acquire object categories and the corresponding linguistic labels without pre-programmed knowledge. His most influential work, "Online Algorithm for Robots to Learn Object Concepts and Language Model" (2016, 23 citations), demonstrates how robots can form conceptual categories and link them to recognized words through continuous interaction, a critical step toward flexible, long-term autonomy. Expanding on this, his paper on "Online joint learning of object concepts and language model using multimodal hierarchical Dirichlet process" (2016, 9 citations) introduces a sophisticated probabilistic model for lifelong learning. By tackling the challenge of practical, long-term concept acquisition, Nishihara’s research directly addresses one of the grand challenges in intelligent robotics: building machines that can learn to use language naturally. His work is foundational for researchers aiming to create robots that understand and communicate about their environment in a human-like, adaptive manner.