lbg-u_5.html

关于自组织神经网络的一种新结构程序,并包含了其它几种神经网络的程序比较

<title>Neural Competitive Models Demo</title>
<strong>
<p align=right><strong>Authors: </strong>
<A HREF="http://www.neuroinformatik.ruhr-uni-bochum.de/ini/PEOPLE/loos">
Hartmut S. Loos</A>
<A HREF="http://www.neuroinformatik.ruhr-uni-bochum.de/ini/PEOPLE/fritzke">
Bernd Fritzke</A>
<hr></p>
<CENTER>

<!--Neural Nets-->

        <H2>DemoGNG (Version 1.5)</H2>
(Please wait while loading ca. 122 KByte class-code.)
<p>
</strong>
</CENTER>

  <i>DemoGNG</i>, a Java applet, implements several methods related to
  competitive learning. It is possible to experiment with the
  methods using various data distributions and observe the learning process.
  A common terminology is used to make it easy to compare one method
  to the other.
<p>

<hr>






<center>
<applet archive="DemoGNG.zip" code="DemoGNG.class" width=720 height=560>
<param name=algorithm value="LBG-U">
<param name=distribution value="Large Spirals">
<b>Your browser does not support Java applets.</b>
</applet>
</center>

<p>
<hr>

<center>
<table border=3 width = 400>
<tr>
<td width = 0>
<td width = 210><center><a href="DemoGNGcode.html"><b>Java Code</b></a></center>
<td width = 210><center><a href="http://www.neuroinformatik.ruhr-uni-bochum.de/ini/VDM/research/gsn/JavaPaper/"><b>Model Description</b></a></center>
<td width = 210><center><a href="tex/DemoGNG/DemoGNG.html"><b>Manual</b></a></center>
<td width = 0>
<tr>
</table>
</center>

  <p>
  Hopefully, the experimentation with the models
  will increase the intuitive understanding and make it easier to
  judge their particular strengths and weaknesses.
  <p>
  The following algorithms are available:
<UL>
<LI> LBG (Linde, Buzo, Gray)
<LI> LBG-U (Fritzke)
<LI> Hard Competitive Learning (standard algorithm)
<LI> Neural Gas (Martinetz and Schulten)
<LI> Competitive Hebbian Learning (Martinetz and Schulten)
<LI> Neural Gas with Competitive Hebbian Learning (Martinetz and Schulten)
<LI> Growing Neural Gas (Fritzke)
<LI> Growing Neural Gas with Utility (GNG-U, Fritzke)
<LI> Self-Organizing Map (Kohonen)
<LI> Growing Grid (Fritzke)
</UL>

If you are not quite familiar with what the applet does, 
have a look at the
<A HREF="tex/DemoGNG/DemoGNG.html"><b>manual in HTML-format</b></A>
or as
<A HREF="tex/DemoGNG.ps.gz"><b>Postscript version</b></A> (ca. 233 KByte).
<p>
For those of you, who want to go into further details:
look at the technical report 
<A HREF="http://www.neuroinformatik.ruhr-uni-bochum.de/ini/VDM/research/gsn/JavaPaper/"><b>Some Competitive Learning Methods</b></A> 
by Bernd Fritzke also available as 
<A HREF="ftp://ftp.neuroinformatik.ruhr-uni-bochum.de/pub/software/NN/DemoGNG/sclm.ps.gz"><b>Postscript version</b></A>.

Furthermore, you can
<A HREF="DemoGNGcode.html#download"><b>Download the whole Package</b></A> or just
have a look at the
<A HREF="DemoGNGcode.html#source"><b>DemoGNG Source</b></A> or the
<A HREF="DemoGNGcode.html#cdoc"><b>Code Documentation</b></A>.