Infopaket

Methods of optimization and metaheuristics

ОК 13

Обов’язкова дисципліна для ОП

Second

2021/2022

2 Semester

5

PR11-Select models and methods and apply hybrid artificial intelligence technologies to solving optimization problems, as well as adjust them depending on the initial data, type of problem and system resources. PR13-Know the classification, main directions in the development of evolutionary modeling systems and metaheuristics; principles of construction and functioning of evolutionary algorithms and metaheuristics; genetic algorithms, evolutionary strategies, genetic programming; basic methods of metaheuristics, intelligent decision-making support technologies based on the evolutionary paradigm and metaheuristics; tools for implementing metaheuristics; classification, properties, main directions of development of evolutionary modeling tools.

Know the basics of linear algebra, mathematical analysis, differential equations, integral and differential calculus, operations research, decision-making methods. Have the skills to work with tools for implementing methods and algorithms, develop appropriate programs in high-level languages, packages for scientific research.

During the study of the discipline, the following substantive sections are considered: classification of approximate methods of combinatorial optimization, constructive algorithms, strategies for developing metaheuristics, combined algorithms, hyperheuristics, exact algorithms, special algorithms, algorithms of deterministic local search, strategies for improving algorithms of deterministic local search, search with pulsating or variable neighborhoods, guided local search, tabu search algorithms, threshold algorithms, stochastic local search algorithms, simulated or simulated annealing algorithms, accelerated probabilistic modeling algorithms (G-algorithms), genetic algorithms (GA), mimetic algorithms, swarm algorithms, ant-algorithms (MA), evolutionary modeling algorithms, automata. Students' training is aimed at solving the tasks of the discipline to form the following competencies: the ability to apply analytical and critical thinking skills to solve problems in the field of computer science; understanding of basic models, methods and algorithms of evolutionary calculations; solving the problem of decision support based on the evolutionary paradigm using metaheuristics in applied fields; development and research of mathematical and evolutionary models, methods and algorithms for solving problems of identification and optimization of complex systems; effective use of metaheuristics in artificial intelligence systems.

1.Zgurovskiy M.Z., Zaychenko Yu.P. The Fundamentals of Computational Intelligence: System Approach. – Springer, 2017. – 395 p. 2.Гуляницький Л.Ф., Мулеса О.Ю. Прикладні методи комбінаторної оптимізації. − К.: «Київський університет», 2016. − 142 с. 3.Снитюк В.Є. Прогнозування. Моделі, методи, алгоритми. − К.: Маклаут, 2008. − 364 с. http://www.twirpx.com/file/1250846/ 4.Poli R., LangdonW.B.,McPhee N.F.A Field Guide to Genetic Programming, 2008. 5.http://digitalcommons.morris.umn.edu/cgi/viewcontent.cgi?article=1001&context=cs_facpubs 6Sean, Luke. Essentials of Metaheuristics (2009). https://yadi.sk/d/6SRXyXaEMxApm 7.Min-Yuan Cheng, Doddy Prayogo. Symbiotic Organisms Search: A new metaheuristic optimization algorithm // Computers and Structures, 139 (2014), P. 98-112.

Lectures, Laboratory classes, Student's work independently

During the semester, students perform laboratory work and present their results to the audience, and two written control works are also conducted. With the prior agreement of the teacher, students can be credited with the results of their learning, confirmed by certificates from the online platform Coursera, as part of their lab work. The maximum number of points a student can receive for their work during the semester is 100 points. The form of the final evaluation is the credit. The assessment is carried out by issuing a final grade, which is defined as the sum of points for all successfully assessed learning outcomes. To receive credit, it is mandatory for students to complete all laboratory work (minimum score - 48 points, maximum - 80 points) and write the control works (minimum score - 12 points, maximum - 20 points). When the resulting final score is 60 or higher, the student is credited.

Ukrainian