Lectures:
1.Current state in the field of soft computing, fuzzy logic, neural networks, evolutionary computing (EVT). Classification of evolutionary computing techniques, historical facts, current trends in the field of EVT. The central dogma of EVT according to Darwin and Mendel.
2. No Free Lunch Theorem. Computational complexity and physical limits of algorithms.
3. Limitations placed on the cost function and parameters of the individual. Penalty and its impact on the geometry of the cost function. Work with real, integer and discrete values of individual parameters.
4. Test benchmark function.
5. Multi and many-objective optimization and Pareto set.
Critical situations in the algorithm run and their solution.
6. Blind search and climbing algorithm.
7. Genetic algorithms. GA terminology. Principle of operation, Hybrid GA, messy GA, parallel GA, migration and diffusion model.
8. Evolutionary strategies. Two-member ES: (1 + 1) -ES. Multipart ES: (μ + λ) -ES and (μ, λ) -ES. Multipart ES: (μ + λ) -ES and (μ, λ) -ES. Adaptive ES.
9. Particle swarm. Scatter Search. AntColony Optimization.
10. SOMA: Self-organizing Migration Algorithm, the principle of operation and used algorithm strategies: ATO, ATR, ATA and ATAA.
11. Differential evolution, principle of operation and used versions: DE / best / 1 / exp, DE / rand / 1 / exp, DE / rand-to-best / 1 / exp, DE / best / 2 / exp, DE / rand / 2 / exp, DE / best / 1 / bin, DE / rand / 1 / bin, DE / rand-to-best / 1 / bin, DE / best / 2 / bin, DE / rand / 2 / bin. SOMA, DE and permutation test problems.
12. Swarm intelligence (SI). Basic terms and definitions, representative SI algorithms - particle swarm, scatter search, ant colony optimization, swarm robotic, artificial evolution of complex systems.
13. Techniques of symbolic regression: genetic programming, grammatical evolution. Alternatives: Analytical Programming, Probabilistic Incremental Program Evolution - PIPE, Gene Expression Programming, Multiexpression Programming and more.
Laboratories (for PC classrooms):
The seminar will focus on the practical application of the discussed techniques and solutions of selected problem examples.
• Implementation of selected benchmark functions (optimization problems)
• Implementation of the Blind Algorithm and the Climbing Algorithm
• Implementation of the Genetic Algorithm and its application to the business traveller problem (TSP)
• Implementation of Differential Evolution
• Implementation of Evolutionary Strategies
• Implementation of the Particle Swarm Optimization algorithm using inertia
• Implementation of a Self-Organizing Migration Algorithm (SOMA), comparison of the behavior of SOMA and PSO algorithms
• Implementation of Firefly algorithm
• Implementation of the Ant Colony Optimization (ACO) algorithm and its application to the business traveller problem. Comparison of ACO and GA in terms of convergence speed and accuracy of the resulting solution
• Teaching-learning based algorithm
• Multiobjective optimization. Implementation of NSGA II algorithm (Non-dominated Sorting Genetic Algorithm)
1.Current state in the field of soft computing, fuzzy logic, neural networks, evolutionary computing (EVT). Classification of evolutionary computing techniques, historical facts, current trends in the field of EVT. The central dogma of EVT according to Darwin and Mendel.
2. No Free Lunch Theorem. Computational complexity and physical limits of algorithms.
3. Limitations placed on the cost function and parameters of the individual. Penalty and its impact on the geometry of the cost function. Work with real, integer and discrete values of individual parameters.
4. Test benchmark function.
5. Multi and many-objective optimization and Pareto set.
Critical situations in the algorithm run and their solution.
6. Blind search and climbing algorithm.
7. Genetic algorithms. GA terminology. Principle of operation, Hybrid GA, messy GA, parallel GA, migration and diffusion model.
8. Evolutionary strategies. Two-member ES: (1 + 1) -ES. Multipart ES: (μ + λ) -ES and (μ, λ) -ES. Multipart ES: (μ + λ) -ES and (μ, λ) -ES. Adaptive ES.
9. Particle swarm. Scatter Search. AntColony Optimization.
10. SOMA: Self-organizing Migration Algorithm, the principle of operation and used algorithm strategies: ATO, ATR, ATA and ATAA.
11. Differential evolution, principle of operation and used versions: DE / best / 1 / exp, DE / rand / 1 / exp, DE / rand-to-best / 1 / exp, DE / best / 2 / exp, DE / rand / 2 / exp, DE / best / 1 / bin, DE / rand / 1 / bin, DE / rand-to-best / 1 / bin, DE / best / 2 / bin, DE / rand / 2 / bin. SOMA, DE and permutation test problems.
12. Swarm intelligence (SI). Basic terms and definitions, representative SI algorithms - particle swarm, scatter search, ant colony optimization, swarm robotic, artificial evolution of complex systems.
13. Techniques of symbolic regression: genetic programming, grammatical evolution. Alternatives: Analytical Programming, Probabilistic Incremental Program Evolution - PIPE, Gene Expression Programming, Multiexpression Programming and more.
Laboratories (for PC classrooms):
The seminar will focus on the practical application of the discussed techniques and solutions of selected problem examples.
• Implementation of selected benchmark functions (optimization problems)
• Implementation of the Blind Algorithm and the Climbing Algorithm
• Implementation of the Genetic Algorithm and its application to the business traveller problem (TSP)
• Implementation of Differential Evolution
• Implementation of Evolutionary Strategies
• Implementation of the Particle Swarm Optimization algorithm using inertia
• Implementation of a Self-Organizing Migration Algorithm (SOMA), comparison of the behavior of SOMA and PSO algorithms
• Implementation of Firefly algorithm
• Implementation of the Ant Colony Optimization (ACO) algorithm and its application to the business traveller problem. Comparison of ACO and GA in terms of convergence speed and accuracy of the resulting solution
• Teaching-learning based algorithm
• Multiobjective optimization. Implementation of NSGA II algorithm (Non-dominated Sorting Genetic Algorithm)