Convolutional Neural Networks (CNN) have been established in recent years as a very promising form of deep learning. Coming from image processing, CNNs have meanwhile entered many areas of deep learning.
The fundamental difference between fully connected and Convolutional Neural Networks lies in the combination of successive levels. In the fully connected case all neurons are connected to neurons of the previous layer. In the Convolutional Neural Networks, however, neurons are associated with a (usually smaller) number of closely spaced next-level neurons. Moreover, all neurons of a layer are connected in the same way to the previous layer. This leads to the operation 'Convolution', which gave this network type the name.
