After successful completion of the course, students are able to 

• train self-organizing maps
• select and combine suitable visualization techniques
• detect patterns in data based on visualizations and generate hypotheses  concerning the cluster structure
• interpret cluster structures and critically reflect on these
• recognize application scenarios for genetic algorithms, cellular automata and ant colony systems
• able to deploy these methods, and understand their parameter optimisation for specific tasks
• describe the functionality of different swarm intelligence systems and assign them to different problem domains based on their strength and weakness 
• implement a particle swarm optimization algorithm for a given problem and tune its parameter accordingly
• understand and reason the basic methodic of swarm robotics as well as its applications 

Unsupervised Learning models, such as self-organizing maps, growing hierarchical structures, cellular automata, ant colony optimisation, ... 

Lecture Dates  winter term 2019/20:

• 10.10.: Introduction (Vorbesprechung)
• 17.10.: Genetic Algorithms, Cellular Automata, Swarms
• 24.10.: Multi-Agent Systems
• 07.11.: Swarm Systems 1
  
• 14.11.: Swarm Systems 2
• 21.11.: SOM 1: Basics, Architecture, Training Process
• 05.12.: SOM 2: Visualizations: class distributions, densities, cluster boundaries, topology violations
• 12.12.: SOM 3: Visualizations (cont.), Quality Measures,  SOM Comparisons, related Architectures

ECTS/Effort:

Lectures: 8 sessions a 2h: 16h

Literature: 10h

Exercises:

    Ex1: 20h

    Ex2: 20h

    Ex3: 20h

Exam Preparation: 25.5h

Exam: 1h

SUM: 112.5h

- Lectures

- Demos and examples discussed during class

- Assignments to be elaborated in small groups

- written final exam

- Submission of writen assignments, partially including oral reviews of these assignment