Machine Learning Fundamentals

Course guarantor:

Vojtěch Franc

Lecturer:

Vojtěch Franc

Tutor:

Vojtěch Franc

Supervisor:

Department of Cybernetics

Synopsis:

The aim of this course is to provide a comprehensive understanding of the fundamental principles underlying machine learning algorithms and to explain their use in basic machine learning algorithms. The goal of statistical machine learning is to design systems incorporating models and algorithms capable of learning to solve problems based on the examples provided and prior knowledge of the problem. This course is designed with two main objectives. First, it seeks to clarify the basic principles of learning, such as risk minimization, maximum likelihood learning, and Bayesian learning, and to delve into their theoretical foundations. Second, it seeks to explore the basic models for classification and regression and show how these models can be effectively learned by applying these basic concepts.

Requirements:

Syllabus of lectures:

1. Gentle introduction to generalization.

2. Empirical Risk Minimization. Statistical preliminaries (convergence of RVs, the law of large numbers, Hoeffding inequality).

3. Generalization theory. PAC learning.

4. VC dimension.

5. Bias-Variance trade-off.

6. Model selection and validation methods.

7. Performance metrics.

8. Linear models. Perceptron algorithm

9. Support Vector Machines.

10. Kernel methods.

11. Deep learning and generalization.

12. Generative learning.

13. Bayesian learning.

Syllabus of tutorials:

Study Objective:

The aim of this course is to provide a comprehensive understanding of the fundamental principles underlying machine learning algorithms and to explain their use in basic machine learning algorithms. The goal of statistical machine learning is to design systems incorporating models and algorithms capable of learning to solve problems based on the examples provided and prior knowledge of the problem. This course is designed with two main objectives. First, it seeks to clarify the basic principles of learning, such as risk minimization, maximum likelihood learning, and Bayesian learning, and to delve into their theoretical foundations. Second, it seeks to explore the basic models for classification and regression and show how these models can be effectively learned by applying these basic concepts.

Study materials:

- T. Hastie, R. Tibshirani and J. Friedman, The Elements of Statistical Learning, Springer, 2010

- M. Mohri, A. Rostamizadeh and A. Talwalkar, Foundations of Machine Learning, MIT Press, 2012

- Y.S. Abu-Mostafa, M.M. Ismail, H.T. Lin. Learning from data. AMLBook, 2012.

- I. Goodfellow, Y. Bengio and A. Courville, Deep Learning, MIT Press, 2016

Note:

Further information:

No time-table has been prepared for this course

The course is a part of the following study plans:

• prg.ai Master (compulsory course in the program)