Deep learning is the recent hot trend in machine learning/AI. It is all about building advanced neural networks. By making multiple hidden layers work in a neural network model, we can work with complex nonlinear representations of data. We create deep learning using base neural networks. Deep learning has numerous use cases in real life, such as, driverless cars, medical diagnostics, computer vision, speech recognition, Natural Language Processing (NLP), handwriting recognition, language translation, and many other fields.

In this chapter, we will deal with the deep learning process: how to train, test, and deploy a Deep Neural Network (DNN). We will look at the different packages available in R to handle DNNs. We will understand how to build and train a DNN with the neuralnet package. Finally, we will analyze an example of training and modeling a DNN using h2o, the scalable open-memory learning platform, to create models with large datasets and implement prediction with high-precision methods.

The following are the topics covered in this chapter:

• Types of DNNs
• R packages for deep learning
• Training and modeling a DNN with neuralnet
• The h2o library

By the end of the chapter, we will understand the basic concepts of deep learning and how to implement it in the R environment. We will discover different types of DNNs. We will learn how to train, test, and deploy a model. We will know how to train and model a DNN using h2o.