Inteligent Algorithms

Teachers in charge

prof. dr. sc. Tomislav Rolich

Study

Textile Science and Technology - postgraduate

Course summary

30

ECTS

4

Knowledge verification

oral exam, Seminar

Lecture type

lectures, seminars, practice

Exercise type

laboratory practice, seminars

Subject content

Biological evolution (the mechanism of evolution, genotype and phenotype, natural selection). Evolutionary algorithms (basic algorithm, individual, population, fitness, mutation). Evolutionary algorithms (recombination, selection, genotype and phenotype, the convergence of the evolutionary algorithm). Evolutionary strategies (basic algorithm, recombination of evolutionary strategies). Evolutionary strategy (1+1)-ES. Multi-membered evolutionary strategy. Selection (the main features of the selection, selection types, static and dynamic selection, elite selection, generational and steady state selection, fitness proportional selection). Selection (fitness function mapping, explicit methods of mapping fitness function, exponential mapping, dynamic linear mapping, windowed mapping). Selection (implicit methods of mapping fitness functions, stochastic sampling with replacement, deterministic sampling, stochastic universal sampling). Recombination (general features of recombination operators, discrete recombination, crossover, one point crossover, one point crossover with permutation, inversion). Recombination (multipoint crossover, uniform crossover, parametric uniform crossover, crossover with more than two parents). Continuous recombination (canonical form of continuous crossover, line crossover, extended line crossover, heuristic crossover, simplex crossover, geometric crossover). Recombination operators in permutation based evolutionary algorithms (partially matched crossover, ordered crossover, cycle crossover, edge crossover, merge crossover).

Aim of course

Make students qualified for inteligent algorithms application to solve practilac optimization tasks. Instruct students in various flavors of inteligent algorithms and application fields

Literature necessary for course

Grundler, D.; Rolich, T.; Rezić, I.: Uvod u evolucijske algoritme, 2012. Grundler, D.; Rolich, T.; Hursa, A.: MATLAB i primjena u tekstilnoj tehnologiji, 2010. Genetic Algorithm and Direct Search Toolbox User's Guide, The MathWorks, Inc., 2015.