Presentation	2001/12/13 Language Acquisition by Robots Naoto IWAHASHI,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	This paper summarizes the methods by which a robot can learn mutual beliefs necessary for language communication with people. The learning is carried out in unsupervised ways based on joint perception and interaction, combining the information of raw speech and visual observations and behavioral reinforcement in probabilistic framework. The beliefs delt with in the methods include phonemes, lexicon, grammar, the influence of behavioral context, and other nonlinguistic belief. In experiments a robot that initially had no linguistic knowledge was eventually able to understand even fragmental and ambiguous utterances according to given situations, and act appropriately. The methods made it possible to reflect the embodied and dynamic aspects of language in the learning process, and they can be extended to provide mre natuoral communication between people and robots.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	robot / language / mutual belief / communication / learning
Paper #	NLC2001-61,SP2001-96
Date of Issue

Conference Information
Committee	SP
Conference Date	2001/12/13(1days)
Place (in Japanese)	(See Japanese page)
Place (in English)
Topics (in Japanese)	(See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To	Speech (SP)
Language	ENG
Title (in Japanese)	(See Japanese page)
Sub Title (in Japanese)	(See Japanese page)
Title (in English)	Language Acquisition by Robots
Sub Title (in English)
Keyword(1)	robot
Keyword(2)	language
Keyword(3)	mutual belief
Keyword(4)	communication
Keyword(5)	learning
1st Author's Name	Naoto IWAHASHI
1st Author's Affiliation	Sony Computer Science Labs.()
Date	2001/12/13
Paper #	NLC2001-61,SP2001-96
Volume (vol)	vol.101
Number (no)	522
Page	pp.pp.-
#Pages	6
Date of Issue