With Artificial Neural Networks (ANN), let's try to simulate typical brain activities such as image perception, pattern recognition, language understanding, sense-motor coordination, and so on. ANN models are composed of a system of nodes, equivalent to neurons of a human brain, which are interconnected by weighted links, equivalent to synapses between neurons. The output of the network is modified iteratively from link weights to convergence.

This final chapter presents ANN applications from different use cases and how neural networks can be used in the AI world. We will see some use cases and their implementation in R. You can adapt the same set of programs for other real work scenarios.

The following topics will be covered:

• TensorFlow integration with R
• Keras integration with R
• Handwritten digit recognition using MNIST dataset with H2O
• Building LSTM with mxnet
• Clustering data using auto encoders with H2O
• Principal Component Analysis (PCA) using H2O
• Breast cancer detection using the darch package

By the end of this chapter, you will have understood the advanced concepts of the learning process and their implementation in the R environment. We will apply different types of algorithms to implement a neural network. We will review how to train, test, and deploy a model. We will look again at how to perform a correct valuation procedure. We will also cover more of deep learning in our use cases as deep learning is the latest thing that is based on advanced neural networks.