An efficient mapping of fuzzy ART onto a neural architecture

Matthias Blume, Sadik Esener

Research output: Contribution to journalArticle

11 Scopus citations

Abstract

A novel mapping of the Fuzzy ART algorithm onto a neural network architecture is described. The architecture does not utilize bi-directional synapses, weight transport, or weight duplication, and requires one fewer layer of processing elements than the architecture originally proposed by Carpenter et al. (1991a). In the new architecture, execution of the algorithm takes a constant time per input vector regardless of the relationship between the input and existing templates, and several control signals are eliminated. This mapping facilitates hardware implementation of Fuzzy ART and furthermore serves as a tool for envisioning and understanding the algorithm.

Original languageEnglish (US)
Pages (from-to)409-411
Number of pages3
JournalNeural Networks
Volume10
Issue number3
DOIs
StatePublished - Apr 1997
Externally publishedYes

    Fingerprint

Keywords

  • fuzzy ART
  • fuzzy ARTMAP
  • neural architecture
  • parallel hardware

ASJC Scopus subject areas

  • Artificial Intelligence
  • Neuroscience(all)

Cite this