Competitive Network

Competitive learning is a form of unsupervised learning in artificial neural networks, in which neurons compete for activation - the right to respond to a subset of the input data. The competition is done using inhibitory connections between neurons in competitive layer. It works by increasing the specialization of each neuron in the network, so each neuron will have maximum response for a particular input pattern. It is well suited for finding clusters within data. [http://en.wikipedia.org/wiki/Competitive_learning]

The dots represent the input vectors, the crosses represent the weights for cach of three units.

To create and train Competitive Network neural network with easyNeurons do the following:

  1. Choose Competitive Network architecture (in main menu choose Networks>Competitive Network)
  2. Enter architecture specific parameters (number of neurons in input layer)
  3. Create training set (in main menu choose Training >New Training Set)
  4. Train network
  5. Test network

Step 1. To create Competitive Network network, in main menu click Networks > Competitive Network

 Step 2. Enter number of neurons in input layer, and click Create button.

This will create the Competitive Network neural network with two neurons in input and four in output layer.

Now we shall train this simple network, to learn from data. First we have to create the training set

Step 3.  In main menu click Training > New Training Set to open training set wizard.

Step 4. Train network

TODO

Step 5. Test network

TODO