Course detail

Artificial Intelligence in Industrial Technologies

FCH-BC_UIPTAcad. year: 2024/2025

After completing the course, students will be able to understand the nature of neural networks and understand the principles of various supervised and unsupervised machine learning models and reinforcement learning. They will gain knowledge of artificial neurons and neural network architecture for solving linearly separable tasks and pattern recognition. This includes both theoretical knowledge and practical skills with the use and training of these algorithms. Graduates of the course will be able to use MLP neural networks as tools for solving practical problems or for evaluating data and optimizing the conditions of the studied process. Given how widespread the use of neural networks and artificial intelligence in general is today, the course student will confidently use this knowledge in his research and in his future employment, both in the field and outside it.

 

 

Language of instruction

Czech

Number of ECTS credits

3

Mode of study

Not applicable.

Entry knowledge

Linear algebra

Basics of statistics, especially univariate analysis (data distribution, symmetry, kurtosis, identification of outliers...)

Matlab Basics – for practical exercises of the subject matter

 

Rules for evaluation and completion of the course

 

Aims

The aim of the course is to provide students with theoretical knowledge and practical skills in the rapidly developing field of Artificial neural networks (ANN), which represent one of the computational models of artificial intelligence (AI). Students will gain experience and get acquainted with applications in the field of signal processing and machine learning (ML, Machine learning), which allows algorithms to learn independently from data without being programmed by humans. Among other things, LTI (Linear time-invariant) systems of the FIR (Finite Impulse Response) type with immutable, optimized and adaptive algorithms with examples of applications in the field of chemistry and chemical technology will be discussed.

 

Study aids

Not applicable.

Prerequisites and corequisites

Not applicable.

Basic literature

Simon S. Haykin, Neural Networks - a comprahensive foundation (2 ed), Prentice Hall, 1999. ISBN 0-13-273350-1 (CS)

Recommended reading

Jure Zupan, Johann Gasteiger. Neural networks in Chemisty and drug design (2 ed), Wiley-VCH, 1999. ISBN: 3-527-29779-0 (CS)

Classification of course in study plans

  • Programme BPCP_MPMU Bachelor's 3 year of study, winter semester, compulsory-optional

Type of course unit

 

Lecture

26 hod., optionally

Teacher / Lecturer

Syllabus

1. LTI systémy I.: Úvod do předmětu, základy LTI (Linear time-invariant) systémů, Matlab.
2. LTI systémy II.: LTI systémy fixní a optimální, úvod do adaptace.
3. LTI systémy III.: Adaptivní LTI FIR systémy standardní a normalizované LMS (Leaking Time-invariant Systems).
4. LTI systémy IV.: Praktické procvičení LTI systémů v Matlabu.
5. Umělé neuronové sítě I.: Úvod do neuronových sítí a strojového učení s učitelem (Supervised learning), bez učitele (Un-supervised learning) a zpětnovazebné učení (Reinforcement learning).
6. Umělé neuronové sítě II.: Artificiální neuron a percepton, vícevrstevné perceptony, aktivační funkce nelineární a prahová, signálové zpracování, klasifikace, segmentace, algoritmy pro trénování, testování modelu.
7. Umělé neuronové sítě III.: Vícevrstevné neuronové sítě s předáváním (FNN, Feedforward neural networks) s iterativním algoritmem minimalizujícím chybovou funkci (Backpropagation algorithm).
8. Umělé neuronové sítě IV.: Praktické procvičení vrstevních neuronových systémů v Matlabu.
9. Neuronové sítě jiných typů I.: Radial basis function (RBF) sítě pro klasifikaci a aproximaci, Rekurentní neuronové sítě (RNN) pro analýzu řeči nebo textu.
10. Neuronové sítě jiných typů II.: Neuronové sítě SON (Self-organizing networks), neuronové sítě se zpětnou vazbou (Feedback Networks), asociativní paměti pro rekonstrukci porušeného nebo neúplného vzoru.
11. Praktické cvičení v Matlabu.
12. Shrnutí probírané látky a aplikace neurálních sítí.