Přístupnostní navigace
E-application
Search Search Close
Course detail
FEKT-MPA-MLRAcad. year: 2024/2025
Students will gain insight into advanced machine learning methods. They will be able to describe and compare the properties of individual approaches to data classification. They will be able to select and apply a specific approach to a given problem. They will also gain practical experience with current implementations of machine learning methods including deep learning.
Language of instruction
Number of ECTS credits
Mode of study
Guarantor
Department
Offered to foreign students
Entry knowledge
- Overview of basic concepts of machine learning.
- Basic knowledge of programming, preferably in Python.
- Mathematical foundations – linear algebra (matrices, vectors), basics of differential calculus and probability.
- Fundamentals of statistics a optimization.
Rules for evaluation and completion of the course
Aims
The course aims to broaden students' understanding of advanced machine learning techniques. Participants will develop the ability to describe, analyze, and differentiate between various data classification methods. They will learn how to effectively select and implement appropriate techniques for specific problems. Furthermore, the course provides hands-on experience with the latest machine learning tools, including deep learning, enhancing their practical skillset in this field.
Study aids
Prerequisites and corequisites
Basic literature
Recommended literature
Classification of course in study plans
Lecture
Teacher / Lecturer
Syllabus
1. Introduction to the issue of classification. Evaluation of classifiers, classification error, testing of classifiers.
2. Feature assessment, selection, and reduction using basic and advanced methods (PCA, mRMR, t-SNE).
3. Linear classifiers – basic principles and methods (perceptron, MSE, SVM).
4. Kernel approach for non-linear classification/regression.
5. Decision and regression trees and forests, random forests.
6. Methods for improving classifier properties (bagging, boosting).
7. Basic principles of artificial neural networks, regularization techniques.
8. Principles of deep learning, deep neural networks (NN) and basic building blocks.
9. Principles of deep NN learning, convolutional NNs, blocks used in deep NNs.
10. Variants of deep NNs, recurrent networks, transformers.
11. Probabilistic models, Methods “Maximum likelihood” and “Maximum a-posteriori probability”.
Exercise in computer lab
1) Introduction to machine learning in pyton, simple classifier example2) Model evaluation methods – metrics, model validation3) Linear and polinomial regression, LASSO/RIDGE regression4) Dual forms and kernels - regression, SVM5) Feature selection - feature filtering, feature wrapping, forward method6) Decision trees, random forest, bias-variance trade-off7) Artificial neural network introduction, introduction to PyTorch8) Introduction to PyTorch 2, simple neural network9) Deep learning in various applications – image classification, signal classification, image segmentation, image2image regression, signal segmentation, signal2signal regression, signal2signal regression10) Transformers - vision transformer example, next word prediction example11) Probabilistic models 1 - Maximum likelihood estimation, Maximum a-posteriori estimation12) Probabilistic models 2 - Naive Bayes Classifier, Gaussian mixture model