annotate toolboxes/FullBNT-1.0.7/nethelp3.3/demrbf1.htm @ 0:e9a9cd732c1e tip

first hg version after svn
author wolffd
date Tue, 10 Feb 2015 15:05:51 +0000
parents
children
rev   line source
wolffd@0 1 <html>
wolffd@0 2 <head>
wolffd@0 3 <title>
wolffd@0 4 Netlab Reference Manual demrbf1
wolffd@0 5 </title>
wolffd@0 6 </head>
wolffd@0 7 <body>
wolffd@0 8 <H1> demrbf1
wolffd@0 9 </H1>
wolffd@0 10 <h2>
wolffd@0 11 Purpose
wolffd@0 12 </h2>
wolffd@0 13 Demonstrate simple regression using a radial basis function network.
wolffd@0 14
wolffd@0 15 <p><h2>
wolffd@0 16 Synopsis
wolffd@0 17 </h2>
wolffd@0 18 <PRE>
wolffd@0 19 demrbf1</PRE>
wolffd@0 20
wolffd@0 21
wolffd@0 22 <p><h2>
wolffd@0 23 Description
wolffd@0 24 </h2>
wolffd@0 25 The problem consists of one input variable <CODE>x</CODE> and one target variable
wolffd@0 26 <CODE>t</CODE> with data generated by sampling <CODE>x</CODE> at equal intervals and then
wolffd@0 27 generating target data by computing <CODE>sin(2*pi*x)</CODE> and adding Gaussian
wolffd@0 28 noise. This data is the same as that used in demmlp1.
wolffd@0 29
wolffd@0 30 <p>Three different RBF networks (with different activation functions)
wolffd@0 31 are trained in two stages. First, a Gaussian mixture model is trained using
wolffd@0 32 the EM algorithm, and the centres of this model are used to set the centres
wolffd@0 33 of the RBF. Second, the output weights (and biases) are determined using the
wolffd@0 34 pseudo-inverse of the design matrix.
wolffd@0 35
wolffd@0 36 <p><h2>
wolffd@0 37 See Also
wolffd@0 38 </h2>
wolffd@0 39 <CODE><a href="demmlp1.htm">demmlp1</a></CODE>, <CODE><a href="rbf.htm">rbf</a></CODE>, <CODE><a href="rbffwd.htm">rbffwd</a></CODE>, <CODE><a href="gmm.htm">gmm</a></CODE>, <CODE><a href="gmmem.htm">gmmem</a></CODE><hr>
wolffd@0 40 <b>Pages:</b>
wolffd@0 41 <a href="index.htm">Index</a>
wolffd@0 42 <hr>
wolffd@0 43 <p>Copyright (c) Ian T Nabney (1996-9)
wolffd@0 44
wolffd@0 45
wolffd@0 46 </body>
wolffd@0 47 </html>