The aim of this course is to teach meta-heuristic methods and evolutionary computation techniques for complex and analytically difficult optimization problems, both theoretically and practically. Students are expected to understand the working principles of nature-inspired algorithms, design suitable solution methods for single and multi-objective optimization problems, and effectively apply these algorithms to real-world problems. The course also aims to develop students' ability to review current scientific literature in the field, critically read and analyze academic articles, and integrate the information they obtain into their own work. Furthermore, it aims to equip students with the competence to generate innovative solutions to engineering problems by understanding the role of meta-heuristic methods in current and applied artificial intelligence processes such as hyperparameter optimization, feature selection, and surrogate modeling.

This course provides a comprehensive introduction to the fundamental structures, variational operators, and optimization processes of genetic algorithms, differential evolution, swarm intelligence-based methods (PSO, ACO, ABC, etc.), and other nature-inspired meta-heuristic algorithms. Evolutionary multi-objective algorithms such as NSGA-II and SPEA2 are discussed in detail for single-objective and multi-objective optimization problems. Furthermore, hyperparameter optimization, feature selection, model improvement strategies, and surrogate modeling approaches are covered within the framework of using meta-heuristic methods in machine learning processes. The course develops students' skills in searching, reading, understanding, and critically evaluating current scientific articles in the field; it also emphasizes the comparative analysis and interpretation of methods presented in the literature. By supporting theoretical explanations with practical applications, example problem-solving, and literature review studies, the course aims to enable students to both develop meta-heuristic solutions suitable for real-world optimization problems and to actively participate in academic research processes.