Название: Modeling Complex Processes Through Nature-Analogous Methods: Artificial Intelligence and Artificial Life
Автор: Christina Klüver, Jürgen Klüver, Jörn Schmidt
Издательство: Springer
Год: 2025
Страниц: 374
Язык: английский
Формат: pdf (true)
Размер: 15.0 MB

This book is an introduction to nature-analogous techniques and related formal methods. For each technique, application examples are provided. It covers cellular automata and Boolean networks, evolutionary algorithms, as well as simulated annealing, fuzzy methods, neural networks, and finally hybrid systems, i.e., combinations of various techniques. Based on the theory of complex dynamic systems, theoretical foundations are also presented, and the similarities of these seemingly very heterogeneous techniques are pointed out. The edition has been revised and expanded with current trends such as ChatGPT.

The current developments under the umbrella term “Artificial Intelligence” are rapid and hardly manageable. While the (hybrid) models we described were a rarity a few years ago, they are now a matter of course. The coupling of methods from Artificial Intelligence (AI), Artificial Life (AL), and Machine Learning (ML) as well as Deep Learning (DL) is currently too diverse to capture the increasing complexity or to process the large amounts of data.

In this edition, the fundamentals continue to be the focus, which we illustrate with comparatively “simple” examples. We refer to current trends that cannot claim to be exhaustive but, in a certain way, continue the concept of the book since the first edition: The methods of AI and AL can be considered in an explicit overall context and can be used both individually and fruitfully combined to solve problems at various levels. This also applies to the algorithms we have developed, whose practical application we also demonstrate in the second edition of the anthology “New Algorithms for Practical Problems. Variations on Artificial Intelligence and Artificial Life.”

The students selected a very challenging problem as well as a model, without prior knowledge in neural networks, Deep Learning libraries, or cloud programming; without ChatGPT, they probably would not have dared the adventure and thus would not have been able to gain valuable experience. This also applies to the students who programmed a CNN in Python with the help of ChatGPT. The computing power also quickly proved to be a problem, which is why Google Colab (or Colaboratory) was used, a cloud-based version of Jupyter Notebook. ChatGPT was used to get explanations for CNN and as an aid for programming in Python.