NEURAL-NETWORK DESIGN USING VORONOI DIAGRAMS

被引：60

作者：

BOSE, NK

GARGA, AK

机构：

[1] Department of Electrical and Computer Engineering, Pennsylvania State University, University Park

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS | 1993年 / 4卷 / 05期

基金：

美国国家科学基金会;

关键词：

D O I：

10.1109/72.248455

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A novel approach is proposed which determines the number of layers, the number of neurons in each layer, and their connection weights for a particular implementation of a neural network, with the multilayer feedforward topology, designed to classify patterns in the multidimensional feature space. The approach is based on construction of a Voronoi diagram over the set of points representing patterns in feature space and this finds added usefulness in deriving alternate neural network structures for realizing the desired pattern classification.

引用

页码：778 / 787

页数：10

共 12 条

[1]

Cybenko G., 1989, Mathematics of Control, Signals, and Systems, V2, P303, DOI 10.1007/BF02551274

[2] HIGHER-DIMENSIONAL VORONOI DIAGRAMS IN LINEAR EXPECTED TIME [J].