Textbook in PDF format This book introduces and explains essential prerequisites for understanding, applying, researching, and further developing the tools currently debated under the terms Machine Learning (ML) and Artificial Intelligence (AI). It strives to be an introductory and comprehensive guide for readers with little prior knowledge, while also offering deeper insights for those interested in advanced aspects and methods beyond the core of the research thread. Overall, this book is intended for anyone seeking a comprehensive understanding of the methods and computer-based applications underlying AI-technology. While digital literacy is beneficial, it is not a prerequisite for understanding the content. Reinforcement Learning (RL) is a widely applied Machine Learning method that in its basics is simple and easy to understand, but in its more advanced variants is a very powerful and flexible method for Artificial Intelligence research. The objective of RL is simply to automatically generate a model, which tells an agent what action to take under what circumstances. The model is generated by repeatedly trying to reach a goal and collecting (digital) rewards for those actions which help bringing the agent closer to its goal and getting (digital) penalties for those actions which distance it from its goal. Basically, this works without any supervision by simply applying brute computational force, meaning that the agent has to be made reaching its goal a large number of times. We will explain the basics of this method on a very simple example task, in which a software agent – the red dot in the image below – has to find the shortest path from each possible position on a 2-dimensional grid – the grey patches – to its goal – the small green house in the center of the grid. Examples in Python. Introduction The use of models in nature – Anticipatory system Reinforcement-learning Evolutionary Computation Machine-based modeling – aka Machine learning Common Machine Learning tools Support Vector Machin k-Nearest-Neighbor Naive Bayes Artificial Neural Networks Natural Language Processing ChatGPT et al. Epilogue: Data ethics References