A perceptron with 2 or more trainable weight layers is called multilayer perceptron or even MLP.It is more powerful than an SLP. A single layer perceptron can divide the input space by way of a hyper plane (in a two-dimensional input space by the use of a straight line). A two stage perceptron can classify convex polygons by furthermore processing these straight lines, e.g. in the form "recognize patterns resting above straight line 1, below straight line 2 and below straight line3. A multilayer perceptron represents an universal function approximator.