A High-Speed Neural Network HS-K-WTA

ZHU Hong; CHEN Qing-hua; LIU Guo-sui

您当前的位置：

首页 >

文章列表页 >

A High-Speed Neural Network HS-K-WTA

更新时间：2025-07-16

- A High-Speed Neural Network HS-K-WTA
- Acta Electronica Sinica Vol. 30, Issue 7, Pages: 1020-1022(2002)
- 作者机构：
  
  1. 南京理工大学电光院,江苏,南京,210094
  2. 南京理工大学计算机系,江苏,南京,210094
  3. 南京理工大学电光院江苏南京,210094
  4. 南京理工大学计算机系江苏南京,210094
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP183
- Published Online：25 July 2002，
  
  Published：2002
- 稿件说明：
移动端阅览
ZHU Hong, CHEN Qing-hua, LIU Guo-sui. A High-Speed Neural Network HS-K-WTA[J]. Acta Electronica Sinica, 2002, 30(7): 1020-1022.
DOI：

ZHU Hong, CHEN Qing-hua, LIU Guo-sui. A High-Speed Neural Network HS-K-WTA[J]. Acta Electronica Sinica, 2002, 30(7): 1020-1022. DOI：

摘要

本文提出一种新的K-Winners-Take-All神经网络:High-Speed-K-Winners-Take-All(HS-K-WTA).HS-K-WTA是以竞争学习算法为基础.HS-K-WTA能够从任何一个数集中

识别出K个较大的数

或较小的数.本文给出HS-K-WTA算法及算法复杂性的分析结果.结果显示HS-K-WTA要比Winstrons更好

更容易硬件实现

更稳定

尤其所取的数集中的数较大时.

Abstract

A new K-Winners-Take-All neural network:High-Speed-Winners-Take-All(HS-K-WTA) is presented.HS-K-WTA can identify the k larger elements( or k smaller ones) in a data set.The analysis results about HS-k-WTA algorithm and its complexity are given.The results show that the speed

the hardware realization

and the stability of HS-K-WTA are much better than Winstrons

especially for a lot of atoms of a data set.

关键词

Keywords

references

Views

952

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

图像分割的新理论和新方法

VLSI Realization of Neural Network

Design of a General Purpose Neuro Processor with Digital-Analog Processing and Discussion on VLSI Integration

Mathematical Principle of Residue-tO-Decimal Conversion by Neural Network

Multi-Valued Neuron（MVN） and Multi-Thresholded Neuron（MTN）,Their Combinations and Applications

Related Author

贾伟宽

史忠植

丁世飞

许新征

Wayne Lang

Hu Jincai

Tan Xilin

鲁华祥

Related Institution

中国矿业大学计算机科学与技术学院

中国科学院计算技术研究所智能信息处理重点实验室

Dept. of Physics, Nanjing Normal University, Nanjing 210024) Wayne Lang

美国北卡罗来纳大学计算机科学系南京 210024

Institute of Semiconductors,Chinese Academy of Science

⁰