Artificial Intelligence

Computational Intelligence and Self-organization (specialization)

This multidisciplinary specialization focuses on the study of organizations, their dynamics and the emergence of different structures for them. The curriculum includes courses from the social sciences and biology.

Relevant courses for this specialization:

• Evolutionary Computing
  The course is treating various algorithms based on the Darwinian evolution theory. Driven by natural selection (survival of the fittest), an evolution process is being emulated and solutions for a given problem are being "bred". During this course all "dialects" within evolutionary computing are treated (genetic algorithms, evolutiestrategieën, evolutionary programming, genetic programming, and classifier systems). Applications in optimisation, constraint handling and machine learning are discussed. Specific subjects handled include: various genetic structures (representations), selection techniques, sexual and asexual genetic operators, (self-)adaptivity. If time permits, subjects in Artificial Life and Artificial Societies, and Evolutionary Art will be handled. Hands-on-experience is gained by a compulsory pogramming assignment.
• Data Mining Techniques
  The course will provide a survey of basic data mining techniques and their applications for solving real life problems. After a general introduction to Data Mining we will discuss some "classical" algorithms like Naive Bayes, Decision Trees, Association Rules, etc., and some recently discovered methods like boosting, Support Vector Machines, co-learning. In the second part of the course a number of most successful applications of data mining will be discussed: marketing, fraud detection, text and Web mining, bioinformatics. In addition to lectures there will be an extensive practical part, where students will experiment with various data mining algorithms and data sets. The grade for the course will be based on these practical assignments (i.e., there will be no final examination).
• Advanced Self-organization
  This course is about the understanding of the behaviour and self-organization of complex systems: systems in which the interaction of the components is not simply reducible to the properties of the components. The general question the we address is: how should systems of very many independent computational (e.g., robotic or software) agents cooperate in order to process information and achieve their goals, in a way that is efficient, self-optimizing, adaptive, and robust in the face of damage or attack? We will look at natural systems that solve some of the same problems that we want to solve, e.g., adaptive path minimization by ants, wasp and termite nest building, army ant raiding, fish schooling and bird flocking, coordinated cooperation in slime molds, synchronized firefly flashing, evolution by natural selection, game theory and the evolution of cooperation. The course includes a practical part in which students implement a simulation of a self-organizing complex system and conduct structured experimental analysis with this simulation.

Further information on the curriculum is available on the website of the Faculty of Sciences:

Computational Intelligence and Self-organization