论文信息 - Active lexicon acquisition based on curiosity

Active lexicon acquisition based on curiosity

It is observed that human infants can successfully acquire lexicon; understanding the relationship between the meaning and the uttered word from only one teaching by caregiver, even though there are many other possible mappings. It is thought that the infants utilize various kinds of cognitive biases for efficient learning. This paper proposes a lexical acquisition model which makes use of curiosity to associate visual features of observed objects with the labels that is uttered by a caregiver. This model is applied to a virtual robot. The degree of curiosity that the robot feels to the objects is determined by the two kinds of saliency; habituation saliency and knowledge one. The former saliency is related to the habituation and the latter one is related to the strength of the association between the visual features and the labels. A robot changes its attention and learning rate based on curiosity. Simulation experiments show that the learning model with curiosity effectively associate the labels with the observed visual features.

M. Asada | M. Ogino | M. Kikuchi

[1] Linda B. Smith,et al. The importance of shape in early lexical learning , 1988 .

[2] K. Pruett. Me, Myself and I: How Children Build Their Sense of Self. 18 to 36 Months. Goddard Parenting Guides. , 1999 .

[3] Luc Steels,et al. Aibo''s first words. the social learning of language and meaning. Evolution of Communication , 2002 .

[4] Naoto Iwahashi,et al. Language acquisition through a human-robot interface , 2000, INTERSPEECH.

[5] Minoru Asada,et al. Cognitive developmental robotics as a new paradigm for the design of humanoid robots , 2001, Robotics Auton. Syst..

[6] Deb K. Roy,et al. Learning visually grounded words and syntax for a scene description task , 2002, Comput. Speech Lang..

[7] Pietro Perona,et al. Selective visual attention enables learning and recognition of multiple objects in cluttered scenes , 2005, Comput. Vis. Image Underst..

[8] Yasuo Kuniyoshi,et al. Inter-modal Learning and Object Concept Acquisition , 2005, MVA.

[9] Takashi Omori,et al. A Computational Model for Taxonomy-Based Word Learning Inspired by Infant Developmental Word Acquisition , 2005, IEICE Trans. Inf. Syst..