Neural Networks Basics samples
Perceptron Classifier [Download]
This sample application represents the very basics and the very beginning of neural network
- single neuron with threshold function, which is known as
perceptron. The application
demonstrates perceptron's usage and learning on the very simple tasks - classification of
data belonging to 2 classes. Of course single perceptron can not solve a lot - it may just
classify data, which are linearly separable. This allows to train it for AND or OR functions.
But does not allow to train it for XOR function, which represents none linearly separable case.
One-Layer Perceptron Classifier [Download]
This sample application is similar to the above one, but it demonstrates classification of
more data classes (also all of them are linearly separable from the rest of data). To be able
to classify more classes this application uses already a layer of perceptrons, but not a single
one. The demonstrated simplest neural network has number of outputs equal to number of classes.
For a given input the network sets one of its outputs to 1 and the rest of outputs to 0. The
output with value set to 1 represents class of the value given to the network.
Delta Rule Learning [Download]
This sample is similar to the above one - it also classifies linearly separable data into
several classes, which means that this sample also demonstrates a layer of neurons. But this
time neurons have continuous activation function, but not a threshold function, which enables
usage of new learning algorithm known as
delta rule learning.