Internal categories with irregular geometry and overlapping in ART networks

Publication Type:

Book Chapter

Source:

Lecture Notes in Artificial Intelligence, Springer, Volume 4177, Santiago de Compostela (ESPAÑA), p.291-300 (2006)

ISBN:

3-540-45914-6

URL:

http://www.springerlink.com/content/237525122365516l/

Keywords:

D.Gomes; M.Fernández-Delgado; S.Barro; neuralcomputing; ART

Abstract:

PolyTope ARTMAP (PTAM) [6] is an ART neural network based on internal categories with irregular polytope (polygon in R^n ) geometry. Categories in PTAM do not overlap, so that their expansion is
limited by the other categories, and not by the category size. This makes the vigilance parameter unnecessary. What happens if categories have irregular geometries but overlapping is allowed? This paper presents Overlapping PTAM (OPTAM), an alternative to PTAM based on polytope overlapping categories, which tries to answer this question. The comparison between the two approaches in classification tasks shows that category overlapping does not reduce neither the classification error nor the number of categories, and it also requires vigilance as a tuning parameter. Futhermore, OPTAM provides a significant variability in the results among different data sets.

AttachmentSize
gomes-lcns06.pdf143.25 KB