2-IKV-115 Introduction to Computational Intelligence

Evaluation during the course: active participation at seminars, presentation

Concluding evaluation: exam

Subject aim: To introduce basic principles of various computational methods of data processing that can can be in sum called computational intelligence. Here we include symbolic artificial intelligence (proposition and predicate logic), nature-inspired computation (connectionist models, evolutionary computation and fuzzy systems), as well as various mathematical and statistical methods suitable for solving problems that require intelligence (from a cognitive agent). After passing the course the student will have gained an insight into the field of computational intelligence enabling him/her to understand basic concepts and algorithms, and will be able to select an appropriate approach for a task at hand.

Brief curriculum of the subject:

1. Introduction to computational intelligence, historical overview of methods and approaches.

2. Architektures of intelligent agents. Structure, components, various representation formalisms. Cognitivism and emergenism.

3. Symbolic artificial intelligence: overview of used methods (of logical reasoning, inferencing).

4. Probabilistic reasoning, Bayesian nets, decision making.

5. Introduction to artificial neural nets (ANN): inspiration from neurobiology, tasks suitable for ANNs.

Feedforward ANNs.

6. Data mining: self-organizing ANNs, feature extraction from high-dimensional data.

7. Recurrent ANNs: temporal structure in data, incorporating time into models.

8. Evolutionary algorithms, optimization.

9. Fuzzy systems, fuzzy logic, linguistic variable, fuzzy reasoning.

10. Summary. Using methods of computational intelligence in cognitive science.

Literature:

Various papers to particular topics.

Language in which the subject is taught: Slovak

Date of the last sheet revision: 6.6.2008