1. SylabUZ
2. Faculty of Mathematics, Computer Science and Econometrics
3. summer term 2023/2024
4. Data Engineering - Second-cycle studies leading to MS degree
5. Machine Learning

Generate PDF for this page

Machine Learning - course description

Aim of the course

The aim of the course is to familiarize students with machine learning algorithms and their proper use in practical issues. After this course the student will be
had the ability to implement machine learning algorithms using specialized Python libraries.

The aim of the course is to prepare students to solve practical problems using both classical statistical models, neural network algorithms and other known machine learning methods. An important goal of this subject is to develop reasoning skills, analytical thinking and selection of appropriate machine learning algorithms for a given problem.

Prerequisites

Save changes

Knowledge of the basics of statistics and programming.

Scope

Lecture/Lab:

1. Introduction to machine learning. Machine learning methods as a decision support technique.
2. Working in the Python environment with libraries containing implementations of machine learning algorithms: Scikit-Learn, Keras, TensorFlow.
3. Data preprocessing and scaling. Cross-validation.
4. Classification of basic machine learning methods.
5. Unsupervised learning. Cluster analysis algorithms: hierarchical clustering, K-means method.
6. Supervised learning techniques. Classification and regression algorithms: linear regression model, logistic model, decision trees, support vector machines.
7. Artificial neural networks.
8. Assessment of the quality of models. Learning curves. Hyperparameter tuning and model regularization methods.

Teaching methods

Lecture: traditional and problem-based, multimedia presentation.

Laboratory: the laboratory program includes deepening the issues discussed during lectures. Solving research problems using learning algorithms
machine using specialized Python libraries. Teamwork. Discussion related to the use of appropriate algorithms and interpretation of intermediate and final results.

Learning outcomes and methods of theirs verification

Outcome description	Outcome symbols	Methods of verification	The class form

Assignment conditions

Lecture. The knowledge acquired during the lecture will be verified by a final test consisting of theoretical questions and practical tasks. The passing threshold is 50% of the total points. The assessment issues on the basis of which the questions will be developed will be provided to students before the exam.

Lab. The skills acquired during laboratory classes are verified on the basis of the student's performance of the indicated tasks and problems assessed in form of reports. The passing threshold is 50% of the total points.

Final grade. The course grade consists of the laboratory grade (60%) and the exam grade (40%). A necessary condition for obtaining a positive final grade is:
obtaining a positive grade in the exam and laboratory

Recommended reading

1. S. Raschka, V. Mirjalili, Python Uczenie maszynowe. Helion, 2019.
2. W. Richert, L. P. Coelho, Building Machine Learning Systems with Python. Pact Publishing, 2013.
3. P. Dangeti, Statistics for Machine Learning: Techniques for Exploring Supervised, Unsupervised and Reinforcement Learning Models with Python and R., Packt Publishing, 2017.
4. Geron Aurelien: Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition, O'Reilly, 2019.
5. Christopher M. Bishop, Pattern Recognition and Machine Learning, Springer, 2007. 

Further reading

1. GoodFellow I, Bengio Y., Courville A. Deep learning. Systemy uczące się, PWN, Warszawa., 2018

Notes

Modified by dr Maciej Niedziela (last modification: 27-04-2024 22:38)