Course detail
Computational Intelligence
FSI-9VIN Acad. year: 2024/2025 Both semester
Computational Intelligence covers a set of nature-inspired computational methodologies and approaches to address complex real-world problems to which mathematical or traditional modeling can be useless. The course introduces basic approaches and advance methods used in the field. Practical use of the methods is demonstrated on solving simple engineering problems. Students will be given time to practice of own optimization tasks.
Language of instruction
Czech
Supervisor
Department
Entry knowledge
The knowledge of basic relations of the optimization and statistics.
Rules for evaluation and completion of the course
Submitting and defence the project which present/uses implementation of selected CI method.
The attendance at lectures is recommended. Education runs according to individual schedules. The form of compensation of missed seminars is fully in the competence of the tutor.
Aims
To give students knowledge of Computational Intelligence fundamentals, i.e. of fundamentals of nature-inspired approaches to solving hard real-world problems. Namely of fundamentals for solving of optimization problems, mathematical models and classification. The various evolutionary algorithms, optimization metaheuristics and artificial neural networks will be presented.
Understanding of basic methods of Computational Intelligence and ability of their implementation.
The study programmes with the given course
Programme D-APM-K: Applied Mathematics, Doctoral, recommended course
Programme D-APM-P: Applied Mathematics, Doctoral, recommended course
Type of course unit
Lecture
20 hours, optionally
Syllabus
The lectures are divided into four blocks:
Block 1: Relationship between Computational Intelligence and Artificial Intelligence. Presentation of engineering tasks. Presentation of student tasks.
Block 2: Evolutionary algorithms, optimisation metaheuristics, swarm intelligence (Genetics Algorithms, Grammatical Evolution, Genetic Programming, Ant Colony Optimisation, metaheuristics HC12).
Block 3: Artificial Neural Networks (feedforward neural networks, recurrent neural networks, self-organisation, deep learning)
Block 4: Individual consultations for own tasks.