Introduction to artificial neural networks

Our natural networks of neurons work as decision-making systems with information processing units called neurons that help us with, for example, recognizing the faces of our friends. Artificial neural networks (ANNs) work similarly. Dissimilar to having a giant network of neurons, as in our bodies, that take care of all decision-making, active or reactive, ANNs are designed to be problem-specific. For example, we have ANNs for image classification, credit risk estimation, object detection, and more. We will use neural networks instead of ANNs for simplicity in this book.

First, we want to focus on fully connected neural networks (FCNNs), which work on tabular data (Figure 12.1). FCNNs and multi-layer perceptrons (MLPs) are used interchangeably in many resources. To be able to better compare different types of neural networks, we will use FCNNs instead of MLPs in this book:

Figure 12.1 – Schematic illustration...