|
Lecturer(s)
|
-
Brandejský Tomáš, doc. Ing. Dr.
|
|
Course content
|
The main aim of this subject is to introduce students into fundabental ideas of evolutionary techniques. Then students can focus to optimization techniqes (and related genetic algorithms and evolutionary strategies, simulated annealing etc.) or to problems of structure development and symbolic regression and thus especially onto genetic programming, grammatical evolution etc. Within the study students will practically test these algorithms. In this subject the attantion to basic knowledge of information dynamics of GA, ES and GPA will be given to allow students efficiently apply these algorithms. The main topics of course: 1.Basic types and structures of evolutionary algorithms, gene reprezentation, evolutionary operators. In the case of genetic programming algorithms study the difference of particular gene reprezentation will be discussed as well as selection of function set and its influence onto efficiency of the algorithm, specific application defined functions and operations, e.g. for solving of problems described by common graphs, hierarchical and hybrid algoritms and their suitability to specific kinds of tasks. 2.Classes of tasks solved by evolutionary techniques, applications in optimalization of non-linear systems, symbolic regression, constructive tasks, programming, game rule discovering etc. 3.Information dynamics of evolutionary algorithms, problem of global optimization, fitness function. 4.Parallel implementation of evolutionary algorithms and suitability of particular classes of parallel HW as multicore and nany core processors, GPGPU, clusters, grids.
|
|
Learning activities and teaching methods
|
|
Monologic (reading, lecture, briefing), Methods of individual activities, Laboratory work
|
|
Learning outcomes
|
The main aim of this subject is to introduce students into fundabental ideas of evolutionary techniques.
|
|
Prerequisites
|
unspecified
|
|
Assessment methods and criteria
|
Oral examination, Written examination, Home assignment evaluation
The student completes at least 3 consultations during the semester concerning the theoretical content of the course. The student will pass at least 1 consultations concerning the assigned practical problem. In practical part student will implement suitable algorithm for given problem solving and will demonstrate its function and suitability. Such algoritm can be e.g. some genetic algorithm, evolutionary strategy for optimization tasks or genetic programming algoritm, hierarchical and hybrid algorithm or algorithm of grammatical evolution for constructive task, eventually some next evolutionary technique chosen with respects to future PhD work of the student.
|
|
Recommended literature
|
-
Golberg, D.E. Genetic algorithms in search, optimization, and machine learning. 1989.
-
Hynek, Josef. Genetické algoritmy a genetické programování. Praha: Grada, 2008. ISBN 978-80-247-2695-3.
-
Koza, John R. Genetic programming III : Darwinian invention and problem solving. San Francisco: Morgan Kaufmann, 1999. ISBN 1-55860-543-6.
-
Poli, R.; Langdon, W.B. A field guide to genetic programming. available at http://www.gp-field-guide.org.uk. ISBN 978-1-4092-0073-4.
|