This page was automatically generated by NetLogo 4.0. Questions, problems? Contact feedback@ccl.northwestern.edu.
The applet requires Java 1.4.1 or higher. It will not run on Windows 95 or Mac OS 8 or 9. Mac users must have OS X 10.2.6 or higher and use a browser that supports Java 1.4. (Safari works, IE does not. Mac OS X comes with Safari. Open Safari and set it as your default web browser under Safari/Preferences/General.) On other operating systems, you may obtain the latest Java plugin from Sun's Java site.
created with NetLogo
view/download model file: nn.nlogo
This is a simple Neural Network. The network learns to convert binary numbers in decimal numbers. A Neural Network is a very simple model of how our brain learns.
The Neural Network must be trained to work. Given the input the answer of the network will depend on the weight of each link between the nodes. To have the right weights you need to train the network.
After the setup you can run the training session. In the graph "errori" you can see the number of errors the network does during the training.
The training consist in the input of 300 example each period (you can change it), the right result is known, so the error can be back-propagated to change the weight of the links between the nodes: the network is learning!
The errors will decrease to zero. When you like you can test the network, just plug in some number as input and control if the answer of the network is right!
You can change the number of examples in each period and change the learning rate, that is how much the network reacts to an error.
Jakob Grazzini 2008.