Naoto Iwahashi
National Institute of Information and Communications Technology, University of Electro-Communications, Sony Computer Science Laboratories, Kyoto University, Kyoto Seika University, Advanced Telecommunications Research Institute International, Okayama Prefectural University, Honda (Japan), Sony Corporation (United States)
Papers
44
Total Citations
906
H-Index
17
About
Naoto Iwahashi is a pioneering researcher in human-robot interaction, symbol emergence, and developmental robotics, whose work sits at the compelling intersection of cognitive science, linguistics, and artificial intelligence. His research addresses one of the most fundamental challenges in robotics: enabling machines to autonomously acquire language and form meaningful conceptual understanding through direct physical and social interaction with their environment. Iwahashi's most influential contribution, "Symbol emergence in robotics: a survey" (2016, 153 citations), provides a comprehensive framework for understanding how robots can develop symbol systems analogous to human language acquisition. Complementing this, his pioneering work on multimodal object categorization and LDA-based word grounding — collectively garnering over 270 citations — demonstrates how robots can integrate audio, visual, and haptic information to form coherent, human-like concepts without explicit supervision. A consistent theme throughout his career is the developmental approach to language learning, reflecting a deep belief that true linguistic competence in machines must emerge organically through embodied experience and semiotic communication. His probabilistic modeling techniques, including multimodal LDA and hierarchical language models, have substantially advanced unsupervised concept formation in robotics. With a body of work spanning over two decades and hundreds of citations, Iwahashi remains an essential reference for researchers pursuing grounded, human-centered artificial intelligence.
Research Focus
Key Achievements
Top Papers
- 1Symbol emergence in robotics: a survey153 citations · 2016
- 2Grounding of word meanings in multimodal concepts using LDA78 citations · 2009
- 3Multimodal object categorization by a robot77 citations · 2007
- 4
- 5
- 6
- 7
- 8Bag of multimodal LDA models for concept formation38 citations · 2011
- 9
- 10