|
Lecturer(s)
|
-
Kebede Zeru Kifle, Ing.
-
Příhodová Kateřina, Ing. Ph.D.
-
Hájek Petr, prof. Ing. Ph.D.
|
|
Course content
|
Biological foundations of artificial and computational intelligence methods. Fuzzy set, linguistic variable, membership function. Fuzzy logic and fuzzy resolution principle. Fuzzy inference systems. Fuzzy expert systems and Fuzzy Prolog. Bayesian networks. Models of neural networks and approaches to their learning. Recurrent neural networks. Deep neural networks. Neural networks with unsupervised learning. Introduction to evolutionary stochastic optimization algorithms, genetic algorithm. Genetic programming, differential evolution, swarm intelligence. Hybrid systems - fuzzy logic neural networks, evolutionary neural networks and evolutionary fuzzy systems. Knowledge and expert knowledge base design. Programming knowledge and expert systems. An example of knowledge management and decision making systems.
|
|
Learning activities and teaching methods
|
Monologic (reading, lecture, briefing), Dialogic (discussion, interview, brainstorming), Laboratory work
- Home preparation for classes
- 28 hours per semester
- Preparation for a credit (assessment)
- 10 hours per semester
- Preparation for an exam
- 30 hours per semester
- Individual project
- 30 hours per semester
- Contact teaching
- 52 hours per semester
|
|
Learning outcomes
|
The aim of the course is to acquaint students with advanced approaches to artificial and computational intelligence and possibilities of their application in the design of intelligent systems.
A student who has successfully completed the course can: explain the biological foundations of artificial and computational intelligence methods; characterize and distinguish probability and fuzzy approaches to uncertainty processing; describe the structure and behavior of fuzzy inference systems; categorize and compare different models of neural networks; characterize evolutionary stochastic optimization algorithms and distinguish individual algorithms; compare the advantages and disadvantages of artificial and computational intelligence methods. A student who has successfully completed the course can: derive knowledge using inference rules of propositional fuzzy logic and first order predicate fuzzy logic; design a fuzzy expert system knowledge base in a logic programming language; design a suitable neural network structure for a given task and teach it on preprocessed structured and unstructured data; design a purpose function and optimize the task using evolutionary stochastic optimization algorithms; design a hybrid intelligent system that takes advantage of the combination of different methods of artificial and computational intelligence. The student who has successfully completed the course is able to: decide independently and responsibly on the basis of a framework assignment and take into account the complexity, constraints and uncertainties associated with that decision; acquire additional expertise, skills and competences in related disciplines; communicate in a clear and convincing way their own professional opinions to public administration professionals and to the wider public.
|
|
Prerequisites
|
unspecified
|
|
Assessment methods and criteria
|
Oral examination, Written examination, Didactic test, Discussion, Class observation
Assignment: participation in exercises (see directive), written tests throughout the semester with a success rate of at least 60 %, completion and successful defence of two projects. The examination is written with a success rate of at least 60 %.
|
|
Recommended literature
|
-
ENGELBRECHT, A. P. Computational intelligence: An introduction. Chichester: John Wiley & Sons, 2007. ISBN 978-0470035610.
-
KRUSE, R. a kol. Computational intelligence: A methodological introduction. London: Springer, 2013. ISBN 978-1-4471-5849-3.
-
NEGNEVITSKY, M. Artificial intelligence: A guide to intelligent systems. Harlow: Pearson Education, 2011. ISBN 9781408225745.
-
RUTKOWSKI, L. Computational intelligence: Methods and Techniques. Berlin: Springer Verlag, 2008. ISBN 978-3-540-76287-4.
|