Machine Learning Methods for Word Learning and Perceptual Categorization

Hansenclever de França Bassani,

Professor Adjunto,

Universidade Federal de Pernambuco,

Centro de Informática - CIn, Departamento de Sistemas de Computação.

lundi 11 mai -- 12h00 - 14h00 -- SU 1550 (Amphitheatre) UQÀM.

Abstract: Concept acquisition is a central ability required for many cognitive tasks, including word learning for language acquisition. There is a significant amount of theoretical work suggesting that certain types of concepts can be learned through the categorization of perceptions. The creation of a computational model for perception categorization, capable of dealing with real-world data, could greatly advance research in the fields mentioned above. It could allow us to replicate in silicon, experiments carried out with human beings and evaluating theoretical models and hypotheses about related phenomena. In this talk, we will describe a perception categorization model that was developed based on state-of-the-art machine learning methods. The proposed model is capable of handling high-dimensional real-world inputs such as image and audio and create categories of perceptions that are consistent with simple concrete concepts expected to be acquired by human subjects with the provided input data. The model has been applied to replicate cross-situational word learning experiments carried out with human subjects in various situations, displaying similar word learning patterns.

References:

• Bassani, H.F.; Araujo, A.F.R., "Dimension Selective Self-Organizing Maps With Time-Varying Structure for Subspace and Projected Clustering," Neural Networks and Learning Systems, IEEE Transactions on , vol.PP, no.99, pp.1,1 (link).
• Bassani, H.F.; Araujo, A.F.R., Dimension Selective Self-Organizing Maps for clustering high dimensional data. In: 2012 International Joint Conference on Neural Networks (IJCNN 2012 Brisbane), 2012, Brisbane. The 2012 International Joint Conference on Neural Networks (IJCNN). p. 1-8. (link)
• Cangelosi, A., 2010. Grounding language in action and perception: from cognitive agents to humanoid robots. Physics of life reviews 7 (2), 139–51.
• Yu, C., Smith, L. B.,  2007. Rapid word learning under uncertainty via cross-situational statistics. Psychol Sci 18 (5), 414–420.
• Yurovsky, D., Yu, C., Smith, L. B., 2013. Competitive processes in cross-situational word learning. Cognitive Science 37 (5), 891–921.
• Trueswell, J. C., Medina, T. N., Hafri, A., Gleitman, L. R., 2013. Propose but verify: fast mapping meets cross-situational word learning. Cognitive Psychology 66 (1), 126–156.
• Harnad, S., 2005. Handbook of Categorization in Cognitive Science. Elsevier Science, Ch. To Cognize is to Categorize: Cognition is Categorization, pp. 20–46.
• Harnad, S., 1990. The symbol grounding problem. Physica D 42, 335–346.
• Cangelosi, a., Hourdakis, E., Tikhanoff, V., 2006. Language acquisition and symbol grounding transfer with neural networks and cognitive robots. IEEE International Joint Conference on Neural Network - IJCNN, 1576–1582.
• Bloom, P., 2002. How Children Learn the Meanings of Words. The MIT Press.