Difference between artificial Neural network, shallow neural network and fed forward neural network.

Artificial Neural Network (ANN): Artificial Neural Network is a deep learning system which is based on neural network. It is composed of numerous interconnected nodes which can store and process information. The information is propagated from input nodes to the hidden layers and then to the output layers. ANNs have multiple layers of nodes which enable them to learn more complex patterns and correlations between data. ANNs are usually trained with a variety of different techniques, such as supervised learning and reinforcement learning.

Shallow Neural Network (SNN): Shallow Neural Network is an artificial neural network that only consists of a single layer of neurons. This type of neural network is much less complicated than deep learning networks and can only be used to classify simple data. While shallow neural networks are limited in their capacity to classify more complex data, they are very useful in quickly analyzing simple data and making predictions based on it. 

Feed Forward Neural Network (FFNN): Feed Forward Neural Network is a type of neural network where connections between the nodes flow in only one direction, from input nodes to the output nodes. Unlike in artificial neural networks where the nodes are interconnected in all directions, feed forward neural networks allow only data to travel from the input to the output without cycles or feedback loops. FFNNs are often used for simple classification tasks as they are much more computationally efficient than ANNs and are able to achieve similar results in less time.

The main key difference between ANN, SNN and FFNN is the number of layers they have. ANNs have multiple layers which can learn more complex patterns while SNNs are composed of only one layer of neurons and are only used to classify simple data. FFNNs, on the other hand, have connections between nodes that flow in only one direction and can be used to quickly classify data without having to train it. All three neural networks have their advantages and disadvantages and should be used depending on the task.