Artificial Intelligence in Astrophysics

Artificial intelligence and its subparts (like evolutionary algorithms, machine learning\(,\ldots \)) are search methods that can be used for solving optimization problems. A particular class of algorithms like bioinspired one mimic working principles from natural evolution (or swarm intelligence) by employing a population-based (swarm) approach, labelling each individual of the population with a fitness and including elements of random, albeit the random is directed through a selection process. In the contemporary astrophysical literature is a lot of research papers, that work with various machine learning methods, while evolutionary/swarm algorithms are almost neglected. In this chapter, we review the basic principles of evolutionary algorithms and discuss their purpose, structure and behaviour. In doing so, it is particularly shown how the fundamental understanding of natural evolutionary processes has cleared the ground for the origin of evolutionary algorithms. Major implementation variants and they are structural as well as functional elements are discussed. We also give a brief overview of usability areas of the algorithm and end with some general remarks of the limits of computing, including demonstration of its use on astrophysical data processing at the end.