Neural Networks

A neural network is one of the most popular machine learning algorithms available to data scientists. It has consistently outperformed traditional machine learning algorithms in problems where images or digital media are required to find the solution. Given enough data, it outperforms traditional machine learning algorithms in structured data problems. Neural networks that have more than 2 layers are referred to as deep neural networks and the process of using these "deep" networks to solve problems is referred to as deep learning. Two handle unstructured data there are two main types of neural networks: a convolutional neural network (CNN) can be used to process images and a recurrent neural network (RNN) can be used to process time series and natural language data. We will talk more about CNNs and RNNs in Chapter 6, Decoding Images and Chapter 7, Processing Human Language. Let us now see how a vanilla neural network really works. In this section, we will go over...