comparison toolboxes/FullBNT-1.0.7/nethelp3.3/demard.htm @ 0:e9a9cd732c1e tip

first hg version after svn
author wolffd
date Tue, 10 Feb 2015 15:05:51 +0000
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:e9a9cd732c1e
1 <html>
2 <head>
3 <title>
4 Netlab Reference Manual demard
5 </title>
6 </head>
7 <body>
8 <H1> demard
9 </H1>
10 <h2>
11 Purpose
12 </h2>
13 Automatic relevance determination using the MLP.
14
15 <p><h2>
16 Synopsis
17 </h2>
18 <PRE>
19 demmlp1</PRE>
20
21
22 <p><h2>
23 Description
24 </h2>
25 This script demonstrates the technique of automatic relevance
26 determination (ARD) using a synthetic problem having three input
27 variables: <CODE>x1</CODE> is sampled uniformly from the range (0,1) and has
28 a low level of added Gaussian noise, <CODE>x2</CODE> is a copy of <CODE>x1</CODE>
29 with a higher level of added noise, and <CODE>x3</CODE> is sampled randomly
30 from a Gaussian distribution. The single target variable is determined
31 by <CODE>sin(2*pi*x1)</CODE> with additive Gaussian noise. Thus <CODE>x1</CODE> is
32 very relevant for determining the target value, <CODE>x2</CODE> is of some
33 relevance, while <CODE>x3</CODE> is irrelevant. The prior over weights is
34 given by the ARD Gaussian prior with a separate hyper-parameter for
35 the group of weights associated with each input. A multi-layer
36 perceptron is trained on this data, with re-estimation of the
37 hyper-parameters using <CODE>evidence</CODE>. The final values for the
38 hyper-parameters reflect the relative importance of the three inputs.
39
40 <p><h2>
41 See Also
42 </h2>
43 <CODE><a href="demmlp1.htm">demmlp1</a></CODE>, <CODE><a href="demev1.htm">demev1</a></CODE>, <CODE><a href="mlp.htm">mlp</a></CODE>, <CODE><a href="evidence.htm">evidence</a></CODE><hr>
44 <b>Pages:</b>
45 <a href="index.htm">Index</a>
46 <hr>
47 <p>Copyright (c) Ian T Nabney (1996-9)
48
49
50 </body>
51 </html>