Hands-On Quantum Machine Learning

Volume 1: Quantum Advantage in Machine Learning - Second Edition

Today, I released the early access version (first chapter) of the second edition of Hands-On Quantum Machine Learning with Python - Volume 1!

You can access it in the “Books” section of pyqml.com.

I wrote the first edition of Hands-On Quantum Machine Learning with Python - Volume 1: Getting Started in 2021. Almost five years have passed since then, which may be a long time for a rapidly evolving field such as quantum computing and quantum machine learning.

However, the reason for this second edition is not simply that technology has changed and the first edition is outdated. In fact, the core content of the first edition is still entirely relevant. The book focuses on the fundamental principles of quantum computing and quantum machine learning, building on Quantum Bayesian Networks and Grover’s algorithm.

These concepts have not changed since their formulation in the 1990s. But two developments since the first edition justify this new version. The first is technical in nature. Qiskit, the quantum SDK used throughout the book, has undergone two major updates and is now available in versions 1.0 and 2.0. These updates brought significant API changes that rendered most of the original code examples unusable.

While these changes could have been fixed with minor adjustments, I decided to revise the entire book more systematically.

The second and more important reason lies in my teaching experience. In recent years, I have been intensively involved in teaching quantum computers and quantum machine learning, which has given me both insights and the confidence to abandon some of the usual habits in this field that I, too, had clung to.

Although the first edition was already practice-oriented compared to many purely theoretical textbooks, I now realize that it still followed a conventional and inefficient teaching pattern: first theory, then equations, then code, finally the concept. This approach starts with the most abstract and difficult material and ends with the most intuitive. This makes learning unnecessarily difficult and obscures relevance.

My experience has shown that reversing this order dramatically changes the learning process. When we start with practice, with code and conceptual understanding, intuition develops first. Mathematics then serves as an explanation rather than an obstacle. So, the correct order is start with the concept, use code, then equations, finally ground in theory.

This reversal transforms confusion into clarity. It shows that quantum computing is not mysterious. It is simply a system defined by logic, structure, inputs, and outputs. To understand it, you don’t need to master quantum mechanics, but simply recognize and explore how the system behaves. This change in teaching philosophy alone justifies a new edition. The goal is not only to correct outdated code, but to introduce a completely new way of learning quantum machine learning.

Another important revision concerns relevance. In the first edition, the problem of predicting the survival of the Titanic served as an introductory example. It was a useful starting point, but it failed to clarify why quantum computing and quantum-based machine learning are important. The new edition addresses this by reframing the book’s focus, which is also reflected in the updated subtitle: Quantum Advantage in Machine Learning. Quantum methods only make sense if they offer an advantage over classical approaches.

This edition therefore addresses a deeper and more timely problem: understanding how our models, algorithms, and even beliefs are shaped, and how cognitive fallacies influence our thinking. Few books on quantum computing or machine learning address this topic directly, yet it may be one of the most important questions of our time.

Read Chapter 1 today.