Deep learning
Deep learning is a Neural Network consisting of a hierarchy of layers, whereby each layer transforms the input data into more abstract representations. These series of layers, between input and output, identify the input features and create a series of new features based on the data, just as our brain. In deep learning the more layers a network has, the higher the level of features it will learn. The output layer combines all these features and makes a prediction. This is different from an Artificial Neural Network. An artificial neural network is only good at learning the weights of a network with one hidden layer but does not contain multiple hidden layers and hence it cannot learn complex features. Deep learning can be expensive and require massive datasets to train itself on. Since, in deep learning, more the neurons (cells in hidden layers) are, the more features it creates, and correspondingly it needs more data to train on. The data and features are exponential related. For example: If you have 10 features then you are required to provide at least 100 data values.