• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site
Bachelor 2022/2023

Machine Learning

Type: Compulsory course (Information Security)
Area of studies: Information Security
Delivered by: Department of Cyber-Physical Systems Information Security
When: 3 year, 1, 2 module
Mode of studies: distance learning
Online hours: 32
Open to: students of one campus
Language: English
ECTS credits: 5
Contact hours: 60

Course Syllabus

Abstract

The discipline "Machine Learning" studies a class of artificial intelligence methods, the characteristic feature of which is not the direct solution of a problem, but learning in the process of applying solutions to many similar problems. The purpose of mastering the discipline "Machine Learning" is to familiarize students with the theoretical foundations and basic principles of machine learning - namely, with the classes of models (linear, logical, neural network), quality metrics and approaches to data preprocessing. Within the framework of the discipline, methods for testing statistical hypotheses, linear regression models, classification and clustering, ensembles and decision trees, neural network technologies of machine learning are studied. The discipline "Machine learning" provides the knowledge necessary for the subsequent passage of undergraduate practice and the preparation of the thesis. During the training, the control of students' knowledge in the form of homework, control, independent work and an exam is provided. Discipline with online course: Calculus and Optimization for Machine Learning - https://www.classcentral.com/course/calculus-and-optimization-for-machine-learning-17335
Learning Objectives

Learning Objectives

  • - familiarization of students with the theoretical foundations and basic principles of machine learning - namely, with the classes of models (linear, logical, neural network), quality evaluation metrics and approaches to data preprosessing;
  • - the formation of students' practical skills in working with data and solving applied problems of data analysis.
Expected Learning Outcomes

Expected Learning Outcomes

  • Knows: concepts and methods of machine learning that can be useful for further study of relevant disciplines, as well as for application in professional activities.
  • Knows: the most popular areas of research in machine learning.
  • Can: choose methods of machine learning for solving problems in the field of professional activity.
  • Owned by: machine learning methods and popular software packages for solving practical problems of machine learning.
Course Contents

Course Contents

  • Introduction to machine learning
  • Statistical estimates and hypothesis testing
  • Machine learning as mathematical modeling
  • Introduction to linear models and the regression problem
  • Linear models and classification problem
  • Selection and evaluation of models, work with features
  • Feature representations for discrete input data
  • Dimension reduction
Assessment Elements

Assessment Elements

  • non-blocking Реализация наивного байесовского классификатора
  • non-blocking Реализация квадратичного дискриминанта и линейного дискриминанта Фишера
  • non-blocking Реализация EM/GMM алгоритма
  • non-blocking Реализация модели линейной регрессии с понижением размерности
  • non-blocking Реализация метода кластеризации k-means и с-means на текстовой информации
  • non-blocking Экзамен
Interim Assessment

Interim Assessment

  • 2022/2023 2nd module
    0.2 * Реализация EM/GMM алгоритма + 0.2 * Реализация модели линейной регрессии с понижением размерности + 0.1 * Реализация квадратичного дискриминанта и линейного дискриминанта Фишера + 0.3 * Экзамен + 0.1 * Реализация метода кластеризации k-means и с-means на текстовой информации + 0.1 * Реализация наивного байесовского классификатора
Bibliography

Bibliography

Recommended Core Bibliography

  • Pattern recognition and machine learning, Bishop, C. M., 2006
  • Вьюгин, В. В. Математические основы машинного обучения и прогнозирования : учебное пособие / В. В. Вьюгин. — Москва : МЦНМО, 2014. — 304 с. — ISBN 978-5-4439-2014-6. — Текст : электронный // Лань : электронно-библиотечная система. — URL: https://e.lanbook.com/book/56397 (дата обращения: 00.00.0000). — Режим доступа: для авториз. пользователей.
  • Флах, П. Машинное обучение. Наука и искусство построения алгоритмов, которые извлекают знания из данных / П. Флах. — Москва : ДМК Пресс, 2015. — 400 с. — ISBN 978-5-97060-273-7. — Текст : электронный // Лань : электронно-библиотечная система. — URL: https://e.lanbook.com/book/69955 (дата обращения: 00.00.0000). — Режим доступа: для авториз. пользователей.

Recommended Additional Bibliography

  • Deep learning, Goodfellow, I., 2016
  • Коэльо, Л. П. Построение систем машинного обучения на языке Python / Л. П. Коэльо, В. Ричарт , перевод с английского А. А. Слинкин. — 2-е изд. — Москва : ДМК Пресс, 2016. — 302 с. — ISBN 978-5-97060-330-7. — Текст : электронный // Лань : электронно-библиотечная система. — URL: https://e.lanbook.com/book/82818 (дата обращения: 00.00.0000). — Режим доступа: для авториз. пользователей.

Authors

  • IVANOV FEDOR ILICH