A neural network for generating diverse decision boundaries with a minimal number of parameters

A neural network for generating diverse decision boundaries with a minimal number of parameters

We present a neural network architecture that can provide a large number of flexible decision boundaries for pattern recognition using a minimal number of parameters. A biologically compatible implementation of the network leads to greater discrimination power.

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Bankman, Isaac N., Sadowsky, John, Sigillito, Vincent G.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Biology
Surface morphology
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present a neural network architecture that can provide a large number of flexible decision boundaries for pattern recognition using a minimal number of parameters. A biologically compatible implementation of the network leads to greater discrimination power.
DOI:10.1109/IEMBS.1992.5761240