From the scientific and philosophical studies conducted over the centuries, special mechanisms have been identified that are the basis of human intelligence. Taking inspiration from their operations, it was possible to create machines that imitate part of these mechanisms. The problem is that they have not yet succeeded in imitating and integrating all of them, so the Artificial Intelligence (AI) systems we have are largely incomplete.

A decisive step in the improvement of such machines came from the use of so-called Artificial Neural Networks (ANNs) that, starting from the mechanisms regulating natural neural networks, plan to simulate human thinking. Software can now imitate the mechanisms needed to win a chess match or to translate text into a different language in accordance with its grammatical rules.

This chapter introduces the basic theoretical concepts of ANN and AI. Fundamental understanding of the following is expected:

• Basic high school mathematics; differential calculus and functions such as sigmoid
• R programming and usage of R libraries

We will go through the basics of neural networks and try out one model using R. This chapter is a foundation for neural networks and all the subsequent chapters.

We will cover the following topics in this chapter:

• ANN concepts
• Neurons, perceptron, and multilayered neural networks
• Bias, weights, activation functions, and hidden layers
• Forward and backpropagation methods
• Brief overview of Graphics Processing Unit (GPU)

At the end of the chapter, you will be able to recognize the different neural network algorithms and tools which R provides to handle them.