School of Computing

An unsupervised dual-network connectionist model of rule emergence in category learning

R.A. Cowell and R.M. French

In Stella Vosniadou, Daniel Kayser, and Athanassios Protopapas, editors, Proceedings of the European Cognitive Science Conference 2007. Taylor and Francis, May 2007.

Abstract

We develop an unsupervised “dual-network” connectionist model of category learning in which rules gradually emerge from a standard Kohonen network. The architecture is based on the interaction of a statistical-learning (Kohonen) network and a competitive-learning rule network. The rules that emerge in the rule network are weightings of individual features according to their importance for categorisation. Once the combined system has learned a particular rule, it de-emphasizes those features that are not sufficient for categorisation, thus allowing correct classification of novel, but atypical, stimuli, for which a standard Kohonen network fails. We explain the principles and architectural details of the model and show how it works correctly for stimuli that are misclassified by a standard Kohonen network.

Download publication 169 kbytes (PDF)

Bibtex Record

@inproceedings{2840,
author = {R.A. Cowell and R.M. French},
title = {{A}n unsupervised dual-network connectionist model of rule emergence in category learning},
month = {May},
year = {2007},
pages = {},
keywords = {},
note = {},
doi = {},
url = {http://www.cs.kent.ac.uk/pubs/2007/2840},
    publication_type = {inproceedings},
    submission_id = {19332_1225983499},
    ISBN = {978-1-84169-696-6 },
    booktitle = {Proceedings of the European Cognitive Science Conference 2007},
    editor = {Stella Vosniadou and Daniel Kayser and Athanassios Protopapas },
    publisher = {Taylor and Francis},
}

School of Computing, University of Kent, Canterbury, Kent, CT2 7NF

Enquiries: +44 (0)1227 824180 or contact us.

Last Updated: 13/07/2009 11:59