# Preface

In just ten years, **Graph Neural Networks** (**GNNs**) have become an essential and popular deep learning architecture. They have already had a significant impact various industries, such as in drug discovery, where GNNs predicted a new antibiotic, named halicin, and have improved estimated time of arrival calculations on Google Maps. Tech companies and universities are exploring the potential of GNNs in various applications, including recommender systems, fake news detection, and chip design. GNNs have enormous potential and many yet-to-be-discovered applications, making them a critical tool for solving global problems.

In this book, we aim to provide a comprehensive and practical overview of the world of GNNs. We will begin by exploring the fundamental concepts of graph theory and graph learning and then delve into the most widely used and well-established GNN architectures. As we progress, we will also cover the latest advances in GNNs and introduce specialized architectures that are designed to tackle specific tasks, such as graph generation, link prediction, and more.

In addition to these specialized chapters, we will provide hands-on experience through three practical projects. These projects will cover critical real-world applications of GNNs, including traffic forecasting, anomaly detection, and recommender systems. Through these projects, you will gain a deeper understanding of how GNNs work and also develop the skills to implement them in practical scenarios.

Finally, this book provides a hands-on learning experience with readable code for every chapter’s techniques and relevant applications, which are readily accessible on GitHub and Google Colab.

By the end of this book, you will have a comprehensive understanding of the field of graph learning and GNNs and will be well-equipped to design and implement these models for a wide range of applications.