annotate toolboxes/FullBNT-1.0.7/nethelp3.3/demev2.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 demev2
wolffd@0 5 </title>
wolffd@0 6 </head>
wolffd@0 7 <body>
wolffd@0 8 <H1> demev2
wolffd@0 9 </H1>
wolffd@0 10 <h2>
wolffd@0 11 Purpose
wolffd@0 12 </h2>
wolffd@0 13 Demonstrate Bayesian classification for the MLP.
wolffd@0 14
wolffd@0 15 <p><h2>
wolffd@0 16 Synopsis
wolffd@0 17 </h2>
wolffd@0 18 <PRE>
wolffd@0 19 demev2</PRE>
wolffd@0 20
wolffd@0 21
wolffd@0 22 <p><h2>
wolffd@0 23 Description
wolffd@0 24 </h2>
wolffd@0 25 A synthetic two class two-dimensional dataset <CODE>x</CODE> is sampled
wolffd@0 26 from a mixture of four Gaussians. Each class is
wolffd@0 27 associated with two of the Gaussians so that the optimal decision
wolffd@0 28 boundary is non-linear.
wolffd@0 29 A 2-layer
wolffd@0 30 network with logistic outputs is trained by minimizing the cross-entropy
wolffd@0 31 error function with isotroipc Gaussian regularizer (one hyperparameter for
wolffd@0 32 each of the four standard weight groups), using the scaled
wolffd@0 33 conjugate gradient optimizer. The hyperparameter vectors <CODE>alpha</CODE> and
wolffd@0 34 <CODE>beta</CODE> are re-estimated using the function <CODE>evidence</CODE>. A graph
wolffd@0 35 is plotted of the optimal, regularised, and unregularised decision
wolffd@0 36 boundaries. A further plot of the moderated versus unmoderated contours
wolffd@0 37 is generated.
wolffd@0 38
wolffd@0 39 <p><h2>
wolffd@0 40 See Also
wolffd@0 41 </h2>
wolffd@0 42 <CODE><a href="evidence.htm">evidence</a></CODE>, <CODE><a href="mlp.htm">mlp</a></CODE>, <CODE><a href="scg.htm">scg</a></CODE>, <CODE><a href="demard.htm">demard</a></CODE>, <CODE><a href="demmlp2.htm">demmlp2</a></CODE><hr>
wolffd@0 43 <b>Pages:</b>
wolffd@0 44 <a href="index.htm">Index</a>
wolffd@0 45 <hr>
wolffd@0 46 <p>Copyright (c) Ian T Nabney (1996-9)
wolffd@0 47
wolffd@0 48
wolffd@0 49 </body>
wolffd@0 50 </html>